Methodology and Data

Calculating the index

This index system measures levels of development in different dimensions. It focuses on the relative position of each province. To derive aggregated field indices and the overall index, data need to be normalized, so all the basic indicators are transformed into a 0-10 score system before aggregation. The scores 0 and 10 indicate the positions of the relevant provinces at the lowest and highest levels of development in certain fields among the 31 provinces (including five ethnic autonomic regions and four municipalities administratively at the provincial level; same below). For positive indicators (greater numbers reflect higher level of development), the scores are calculated using the following equation:

For negative indicators (smaller numbers reflect a higher level of development), the following equation is used:

The transformed indicator forms a sub-index or a second grade sub-index. Aggregation of a few sub-indices forms a field index. The total of 10 field indices form the overall index.

Weights of field indices and sub-indices

An important issue is how to determine the weight of each field index in order to construct the overall index, or weights of each sub-index for a field index. There are alternative ways to determine the weights:

1. Weighting based on the judgment of analysts or experts;

2. A Principle Component Analysis Method may be used for weight determination; or,

3. Simple average.

Method 1 is not adopted in this study because it is more or less arbitrary. Previous experiences show that, when the number of indicators included is relatively large, methods 2 and 3 usually lead to very similar results. An additional consideration is that when the indices are updated, the Principle Component Analysis Method leads to changes of the weights. This has the side effect that indices become incomparable over years. However, maintaining comparability is very important when an index is regularly updated. Meanwhile, the simple average method can give consistent measures from year to year, so that changes in the level of development in each province can be correctly traced. For this reason, a simple average method is adopted. This means that all field indices are equally weighted in constructing the overall index; and all sub-indices in one field are equally weighted to form a field index.

In general, the indices show the relative position of development of each province, in particular relate to the provinces with the best or least performance overall and in different fields. This is measured with scores and ranks.

The relative performance of each province is also indicated by its rank among all 31 provinces. Finally, their performance in different fields can be traced by the basic indicators, which provide the original information on the absolute achievements of each province without comparison.

The index can be updated regularly when new information is available. Thus changes in the development of the provinces over years can be traced.

Data for RDIC are mainly cited or calculated from the latest available statistics, mostly from the National Bureau of Statistics of China (NBS, various years, 2005b). A few indicators are provided by related government departments or from sample surveys (Fan, Wang and Zhu, various years). Statistical data are mainly for 2005; where they are unavailable, earlier data are used.

Download this Discussion Paper [ PDF 222.9KB| 32 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.