Indonesian Higher Education 127
TABLE 5.6
Multinomial Logit Estimates of School Choice ModelVariables Private NR Private Islam Private Christian
b/(t) b/(t) b/(t)
MALE 0.335* 0.197 0.360
(0.111) (0.146) (0.295)
ISLAM 0.449* 2.349 –1.982
(0.185) (0.579) (0.316)
LANGINDO 0.405 –0.417* –0.695*
(0.115) (0.160) (0.326)
URBAN 0.438 0.043 0.500
(0.129) (0.156) (0.350)
SIBLING 0.004 –0.040 –0.085
(0.042) (0.057) (0.115)
WORKSMP 0.221 0.036 0.662
(0.248) (0.336) (0.589)
PRIFAIL 0.898 0.767 0.834
(0.164) (0.207) (0.404)
NEMSMP –0.025 –0.011 –0.024
(0.005) (0.005) (0.012)
PRIVATENRJH 2.161** 1.110 0.849
(0.186) (0.281) (0.520)
PRIVATEISJH 1.079 .353 0.285
(0.185) (0.181) (0.754)
PRIVATECPOJH 1.261 –0.128 3.608
(0.385) (1.031) (0.431)
FATHSHHE –0.582 –0.615 –0.202
(0.159) (0.231) (0.407)
MOTHSHHE –0.371* –0.417 0.135
(0.204) (0.312) (0.550)
CONSTANT –1.100 –3.525 –1.402**
(0.275) (0.599) (0.611)
Pseudo R-Square 0.189
N 12101
Chi2 1 714.333***
Notes: Omitted category is public school group. T-statistics are in parenthesis and heteroscedasticity
consistent.
- Significance at 10 per cent level, ** Significance at 5 per cent level, and *** Significance at 1 per cent
level. Dummies representing the province where individual was lived were also included in the model.
attend private schools than public schools. Students who help finance the
family by working when attending lower secondary school have lower
odds to attend public school. This result suggests that students from poor