136 Mohamad Fahmi
difference on higher education participation between public and private
Islam. Those determinants are, from the largest to the smallest, earning
foregone (INCOME), parents’ education background (PARENTS), personal
characteristics (PERSONAL), and academic performance (ABILITY). All
gaps are significant at level 1 per cent. The other two gaps, school quality
and family background (FAMILY), are small and not significant.
Similar to the result on private secular decomposition, INCOME is
the highest contributor to the difference between public and private Islam
schools; it contributes 0.067, or 6.7 percentage points gap. Also similar to
private secular decomposition, parents’ education contribute significantly
to the gap; FATHSSHE contributes 0.025 which is higher than MOTHSHHE
that contributes 0.013. The other big gap is contributed by PERSONAL
or personal characteristics. LANGINDO is the biggest contributor and
significant in PERSONAL. However, unlike the private secular gap,
LANGINDO has a positive contribution as it is evident that socio economic
background has an important role in the gap between public and private
Islam school. The last significant gap is the difference in ABILITY. The
gap on ABILTY mostly comes from the normal score of national test or
ZNEMSMA (0.007) that is significant at level 1 per cent. Again, similar to
the private secular decomposition, the large contribution in ABILITY is
primarily due to large gap in the average score; the average of national
test score of public school is 33.175 whereas the average score of private
Islam students is only 29.443.
PARENT is the second strongest determinants that contribute to the
gap between public and private Islam school. It contributes 0.074 as only
two variables are significant statistically; FATHSHHE and MOTHSHHE.
The share of FATHSHHE to the gap is 0.038 whereas that of MOTHSHHE
is 0.030. The strong influence of FATHSHHE and MOTHSHHE to the
gap is expected since mean of FATHSSHE and MOTHSSHE of private
Islam group is lower than public school. The mean of FATHSHHE and
MOTHSHHE in private Islam are 0.282 and 0.194 whereas in public school
they 0.474 and 0.308 respectively.
The third determinant that statistically significantly contribute to the
gap between public and private Islam group is FAMILY as it is significant
at 5 per cent. Despite the significance of FAMILY, all variables in FAMILY
are not statistically significant. LANGINDO and SIBLIG contribute lower
than 0.000 to the gap whereas URBAN and WORKSMA contribute 0.003
and 0.006 respectively. Moreover, SCHOOL or school quality gap is small