Appendix 5: Percentage of Boarding Students Receiving Living
4.3 The Method to Measure the Equity
For easy depiction, we usexto represent an education resource (such as per-student educational expenditure),ito stand for individual or analysis unit (such as student, schools, county), andwto represent the endowment or background variables (such Table 4.1 The indexes and specific indicators of fund input
Index Specific indicator (Yuan per year) Education
revenue
Per-student average education revenue (excludingfiscal appropriation for construction)
Per-student averagefiscal appropriation Per-student average non-fiscal appropriation Regular
expenditure
Per-student average regular expenditure
Per-student average budgetary regular expenditure Per-student average non-personnel expenditure Per-student average current non-personnel expenditure Teacher’s salary Per-teacher average salary
Physical facilities
Per-student averagefloor space of schoolhouses (m2) Per-student average special equipment value Per-student average books (book)
as revenue). We hereby give the meanings of these measurement methods. For technical details, please refer to Appendix.
4.3.1 Horizontal Equity
The method of horizontal equity measurement focuses on the differences of edu- cation resources distribution among groups or individuals. Horizontal equity takes absolute equity as its pursuit. Absolute equity means that different groups or individuals should be provided with the same education, especially fund inputs.
There are a handful of indexes developed by the scholars to measure horizontal equity. These indexes and their characteristics are shown in Table 4.2. Those characteristics may explain the equity or inequity from various perspectives. In this study, we only chose the Gini coefficient and Theil coefficient measure and describe the degree of the horizontal equity.
The Gini coefficient is often used because it is able to be decomposed based on revenue sources and time. The former tells us what source takes the most important role in the Gini coefficient; the latter may tell us what causes the change in the Gini coefficient. These are a concern of ours and policy makers as they are helpful to design a betterfinancial policy for compulsory education.
The Theil coefficient can let us know where the total difference is from—the intra-group difference or the inter-group difference. For example, 60 % of the total
Table 4.2 The indexes of horizontal equity and their characteristics
Method Characteristics
Range ratio The ratio of maximum to minimum; impossible to measure the difference in observation unit used between the maximum and minimum values
Quantile difference May measure the differences of various values at distribution points Relative mean
deviation
Has no sensitivity to the transfer from lower to higher levels along the same side of the mean value
Mcloone index Used for analysis of differences below the median Verstegen index Used for analysis of differences above the median Coefficient of
variation
Has strong sensitivity to anyxtransfer taking place horizontally but is not influenced by inflation
Logarithm standard deviation
Is more sensitive to thexchanges on the lower levels and thus highlights the differences ofxat lower level
Gini coefficient The most direct measurement ofxdifference, gives consideration to the difference ofxbetween each twoi, and is not disturbed by inflation
GE index and Theil coefficient
The GE index, from the low to high value of parameters, changes from more sensitive to the differences on the low level to the differences on the high level
Atkinson index Has a monotonic transformation relationship with GE index
difference is from the inter-group difference. Policy makers and decision makers should try to eliminate the inter-group difference in order to reduce the total difference.
In the measurement of horizontal equity, a problem is whether the criteria to judge the inequity are reasonable and acceptable. Odden and Picus (2000) gave a judgment criteria based on the experience in the United States: the Gini coefficient should be equal to or less than 0.05.
4.3.2 Financial Neutrality
Methods for measuringfinancial neutrality are based on the relationships between x(dependent variable, fund input) and w (independent variable, individual back- ground). The basic assumption is as follows: the closer the relationship between them is, the higher the inequity is. In education finance, regression analysis is generally used to measure such a relationship. The specific measuring methods are divided into four categories: correlation coefficient, slope, elasticity, and adjusted relationship measurement. These methods may be classified into 11 types of specific techniques (see Appendix).
In this report, the correlation coefficient and elasticity analysis are mainly employed to measurefinancial neutrality. The correlation coefficient, whose value is from 0 to 1, is usually used to measure the relationship betweenxandw; here, 0 indicates the highest degree of financial neutrality. The higher the correlation coefficient is, the lower the financial neutrality is. Elasticity is the coefficient of wfrom the simple regression of xon w. If the coefficient of wis significant and greater than 0, we can say thatfinancial neutrality does not exist. According to the experience of some countries, if the correlation coefficient is less than 0.5 and the elasticity or the coefficient ofwis less than 0.1, the educationfinance system in a region satisfies thefinancial neutrality criteria (Odden and Picus2000).
In this study,financial neutrality means that all the governments should provide the same education for every child, whether they are located in rich or poor areas with good or badfinancial ability. In other words, the education a child receives should not be determined by the local economy, local treasury, or local govern- ment’sfinancial ability. Based on much research, we use per capital gross domestic product (GDP) to stand for the local economy and local treasury and use per capital fiscal revenue to reflect thefinancial ability.
4.3.3 Vertical Equity
Before we measure the degree of vertical equity, we must answer three questions:
1. How do we define different backgrounds?
2. How does the government provide different education or fund inputs for groups with different backgrounds? In educational finance, this means how a varied
treatment is performed for groups with different backgrounds in education resource distribution.
3. Do we have any method to measure vertical equity?
In this study, we define the different background as rural and urban areas. The children and their schools in rural areas are disadvantaged groups, and we think that the government should input more funds into rural areas than urban areas. Namely, the government should provide a better education or more education resources for those disadvantaged groups. However, what excess of disadvantaged over advan- taged groups is reasonable? We cannot answer this question. However, we hope to see that the difference between rural and urban areas decreases gradually in the twenty-first century.
The next step is to measure the difference between the two groups. In this study, we mainly adopt three methods examine the differences between rural and urban areas: the absolute difference of the two groups, the ratio of per-student education resources, and the comparison of growth rate of per-student education resources.