The difference in utilities between the best available public school and the best available private option is:
4 .
ln ln
ln ln
U Y
P Y
P Y
T Y
T T
T O
O
pub pri
pub pri
pub pri
pub pri
1 1
2 2
= -
- -
+ +
- +
+ -
+ -
d d
v c
D a
a a
a a
a k
k k
k k
k Under the assumption that P
pri
P
pub
, it can be shown that: 5
, U
0 and 2
2 v
D 6
. Y
U
1 5
2 2
D This model confirms the intuition that a parent that places higher value on educa-
tion is more likely to send their child to public schools, which are assumed to be of higher average quality. Meanwhile, holding other factors constant, wealthier house-
holds are more likely to send their children to private school. In the United States, where private schools are generally considered to be higher quality, wealthier, and
more motivated students tend to select into private schools. In Indonesia, however, where in general public schools are considered to be of higher quality, the two sources
of selection bias are of opposite sign. Thus, the direction and magnitude of bias in the OLS models is unknown, and depends on the relative strength of unmeasured
wealth and unmeasured motivation in determining students’ choice of school and their test score.
IV. Data
The primary data source for this study is the three full rounds of the Indonesia Family Life Survey IFLS1, IFLS2, and IFLS3. See Frankenberg and
Karoly 1995; Frankenberg and Thomas 2000; Strauss et al. 2004. The first round of the survey sampled 7,200 households in 1993; subsequent surveys attempted to rein-
terview these households and households to which previous household members had moved. The 1993 sample was drawn from 321 randomly selected villages, spread
among 13 Indonesian provinces containing 83 percent of the country’s 200 million people. The 321 villages were selected from the sample frame of the 1993 SUSENAS,
the national economic survey, and are located in 149 districts. The sample captures an impressive amount of Indonesia’s remarkable ethnic and geographic diversity.
We analyze the national Ebtanas test scores of former junior secondary school stu- dents. In Indonesia, students are evaluated at the end of the three main education lev-
els elementary, junior secondary, and senior secondary on the basis of a national test.
6
Data on test scores at all three levels were collected in 1997 and 2000 from all The Journal of Human Resources
534
5. The proof is given in Appendix 1. 6. The Ebtanas results are not officially a prerequisite for certification or accreditation purposes but the exams
may be used as a criterion for selecting students. These multiple-choice tests are designed by the Examination Center EC, under the authority of the Office of Educational and Cultural Research and Development in the
Ministry of National Education. EC manages a test item bank from which it draws questions to be included
household members between the ages of 14 and 25. The survey also ascertained the type of school attended at each level.
This study utilizes data on the presence of public and private schools at the district level to identify the effect of school type on student’s test score.
7
District-level data on the presence of schools come from the 1998 round of annual census of schools
conducted by the Indonesian Ministry of Education. Eighty percent of the 42,000 sec- ondary schools in Indonesia responded to this survey. Unfortunately, because of a
budgetary shortfall during the 1998 financial crisis, the education census does not contain any useful information about private schools, except for their private status
and location. This information is used to construct both the total number of junior sec- ondary schools and the percentage of district junior secondary schools that are public
in the district.
The sample consists of all students who reported, in either 1997 or 2000, taking the junior secondary school test between 1990 and 2000.
8
Of the 5,437 respondents that reported taking the national junior secondary school exam between 1990 and 2000,
605 were dropped because they did not report scores from their junior secondary school test. 335 other respondents were dropped because they not report their ele-
mentary school test score, which is a highly significant determinant of junior high test score. An additional 115 respondents were not included in the sample because they
did not report the district of the junior secondary school they attended. The remaining full sample consists of 4,382 respondents. The 1,055 excluded respondents are
slightly less likely to have well-educated mothers and fathers, and somewhat less likely to have attended public school. This suggests that attrition bias, if anything,
might lead to a slight understatement of the positive effect of public schools on test score.
V. Empirical Framework
