goods creation and social transfers may also have a preference for creating a high quality
public education system, equalizing incomes and earning prospects of the future generations.22
If, however, educational expenditures or education quality have a positive effect on student
achievement, their exclusion will positively bias the effect of government size in our previous
models since the variables are likely to be positively correlated.23
Table 8. Post-communist period and school quality measures
|
(1) |
(2) |
(3) |
(4) |
(5) |
(6) |
(7) |
(8) |
(9) | |
|
Gov. consumption |
-0.856 (0.569) |
- |
- |
-0.924 (0.637) |
- |
- |
-0.587 (0.550) |
- |
- |
|
Gov. social expenditures, Income tax rate progressivity GDP per capita (log) |
- - 0.554 |
-1.360* - 0.659 |
- -0.093+ (0.054) |
- |
-1.496** - 0.152 |
- -0.116* |
- - 1.347* |
-1.394** - 1.216 |
- -0.132** |
|
Percentage secondary school Population size (log) Primary education per pupil, percent of GDP (log) |
(0.577) -0.498 (0.717) 2.836+ |
(1.008) (1.077) |
(0.676) -1.039 (0.878) 1.579 |
(0.616) -1.168+ (0.646) 1.178 |
(1.024) -0.168 (0.868) 6.258 |
(0.697) -1.426+ |
(0.542) (0.533) |
(1.030) (0.794) |
(0.618) -0.166 (0.725) |
|
(1.557) - |
(4.117) - |
(1.771) - |
(1.516) (0.295) - |
(4.087) (0.341) - |
(1.466) (0.287) - |
(1.190) -0.005 |
(3.189) 0.048 |
(1.363) -0.003 | |
|
Sample |
All |
OECD |
All |
All |
OECD |
All |
All |
OECD |
All |
|
Time period 1990 - 2003 |
yes |
yes |
yes |
no |
no |
no |
no |
no |
no |
|
Country fixed effects |
yes |
yes |
yes |
yes |
yes |
yes |
yes |
yes |
yes |
|
Time fixed effects |
yes |
yes |
yes |
yes |
yes |
yes |
yes |
yes |
yes |
|
Observations |
177 |
102 |
163 |
166 |
104 |
152 |
185 |
103 |
160 |
|
Countries |
58 |
28 |
56 |
52 |
27 |
50 |
58 |
27 |
55 |
|
R2 |
0.9617 |
0.8791 |
0.9626 |
0.9558 |
0.8740 |
0.9609 |
0.9484 |
0.8833 |
0.9645 |
|
R2 (within) |
0.2251 |
0.3290 |
0.1744 |
0.2377 |
0.3010 |
0.3161 |
0.2462 |
0.3398 |
0.2094 |
|
F-test (social spending, school |
1.5582 |
3.749 |
4.3742 |
0.5928 |
5.2155 |
3.9535 | |||
|
(p-value)______________________ |
0.2156 |
0.029 |
0.0155 |
0.5545 |
0.0081 |
0.0226 |
Note. Absolute t-values in parentheses, + * and ** denote significance at 10, 5 and 1 percent level, respectively.
22 In principle, the quality of teaching may also be influenced by the government creating incentives for teachers.
That teachers may influence student effort is discussed in the model section.
23 The correlation coefficients of per pupil spending in primary education with our government consumption and
social spending exceed well 0.5, while those with pupil-teacher ratio in primary education are -0.76 and -0.29,
respectively.
24