# Table 3 Determinants of number of partners in the past 12 months for men.

REGRESSION MODEL
(1) Y = N (2) Y = √N (3) Y = ln(N+1)
Covariable β se(β) β se(β) β se(β)
Age (years) n.s.   n.s.   n.s.
Square root of age n.s.   n.s.   n.s.
African American 1.058 .321 .306 .077 .233 .055
Mexican American n.s.   n.s.   n.s.
Regular Partner -1.107 .253 -.214 .058 -.191 .043
Overweight -.691 .248 n.s.   n.s.
Obese n.s.   -.198 .062 -.148 .046
Height n.s.   n.s.   n.s.
Income >20,000\$ n.s.   n.s.   n.s.
N of weekly drinks n.s.   n.s.   n.s.
Current smoking n.s.   n.s.   n.s.
Education n.s.   n.s.   n.s.
Constant 2.684 .249 1.237 .052 .923 .038
1. n.s. = not selected by stepwise regression
2. Stepwise regression (p entry .05, p removal .10) of number of partners in the last 12 months, and the root number of partners in the last 12 months on selected covariables. The β coefficients represent the increase in the outcome (dependent) variable (e.g. number of partners N) for every unit increase (e.g. one year, for age) of the covariable to which it belongs. Thus, regression model (1) below can be expressed as N = 1.058.I(African American) - 1.107.I(has a regular partner) -0.691.I(BMI>=25)+2.684, where I(condition) = 1 if the individual meets that condition (e.g. the individual is an African American) and = 0 otherwise.