Page 392 - Contributed Paper Session (CPS) - Volume 2
P. 392
CPS1899 Ira Joy H.M.
and interpreted in relation to the use of drugs. All analysis will be generated
using STATA 13.
3. Result
A total of 290 independent variables was tested for association with the
dependent variable, have or have not tried using drugs, using Rao-Scott.
Among these tested variables, 152 were significantly associated with drug use.
These variables are checked for multicollinearity then regressed to the
independent variable.
Table 1 shows the coefficients, standard errors, p-values, and odds ratios
of the fitted model regressors. There were a total of 16 significant predictors
for drug use.
Table 1. Estimated coefficients, Odds ratios and p-values in weighted
binary logistic model of drug use (Have ever used drugs)
Predictors Coefficient Odds Ratio
Single Male 1.7028 5.4894
Married Male 1.9523 7.0449
From Region 7 1.2486 3.4854
From Region 10 1.4388 4.2158
From Region 11 1.8654 6.4586
From Region 12 0.8888 2.4322
From ARMM 1.3380 3.8114
Often attends religious services (at least once a week) -0.6080 0.5445
From birth up to age 18, 2 person approves having a 0.6342 1.8855
nd
bf/gf before age 18
Family members frequently or almost always eat -0.1911 0.8260
together at least one meal a day
Sees gambling as wrong -0.3696 0.6910
Rated Satisfied with life, all things considered -0.2915 0.7471
Member of a fraternity/sorority 1.3429 3.8301
Had been suspended from attending classes 0.9211 2.5122
Have ever worked for pay, in cash or in kind 0.8068 2.2406
Had ever read pornographic or sexually explicit 0.9926 2.6982
books/magazines/tabloids
constant -5.8863 0.0028
**base category of the dependent variable: Never used drugs
381 | I S I W S C 2 0 1 9