Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 General Research Organization and Contents

1.1 Research Framework

GEC research is a whole new area; there is neither mature research model or methodology, nor existing research contents for reference. Therefore, such researches need to first summarize the related previous researches and then extend the research with more contents and innovation in methodology. GEC is a cross-discipline research involving multiple areas of environmental economics, biology, economics and sociology, which are intricately interrelated; at the same time, we need to rationally define and objectively evaluate GEC, and make innovations in methodology. In this way, we can thoroughly explore the inherent essence of GEC and reveal the laws of GEC evolution. For such a complicated research subject, it requires clear organization and correct research approach, strictly following the designed technical roadmap (See Fig. 5.1) to ensure satisfaction of research standards and quality.

Fig. 5.1
figure 1

GEC research technical roadmap

Full size image

In terms of contents, through substantial literature review and reference to theories about environmentology, economics and management science, the significance and necessity of GEC research is profoundly discussed from different angles; the key contents of this research is to construct the GEC theory system based on the results of previous international researches. Particularly, as a new research area, how to define the term of GEC and how to summarize the characteristics, component factors and motive power of GEC, are the focus and challenge of this study.

As to evaluation methodology, competitiveness research can’t be done in separation with evaluation, which requires construction of scientific and objective evaluation model. Any evaluation model and method shows certain degree of subjectivity and orientation, and the contents reflected by such subjectivity and orientation might have certain guiding effect on the development and formation of the evaluated target. GEC evaluation model not only borrows and applies the mainstream methodology for international competitiveness study, but also shows unique features related to the characteristics of GEC; it must be able to objectively evaluate the GEC of all countries and reflect the internal mechanism and key points of GEC; the evaluation results should also adequately reflect the philosophy behind the research, which is good to the course of global environmental protection and development of ecological economy and good to the realization of global sustainable development. The evaluation model mainly includes two parts, factor model and indicator system; the former uses quantitative analysis to conduct empirical test on the factors influencing GEC based on the connotations and characteristics of GEC, providing reference to analysis on the driving force of GEC, as the foundation of the GEC evaluation indicator system. The indicator system is the basis of competitiveness evaluation and construction of an indicator system that scientifically and objectively reflects the connotations of GEC is a very important part of GEC research. Selection of the indicators is not random and must follow definite principles; they are screened out using hierarchical model and weighted according to certain methods after careful investigation. In terms of evaluation methodology, the most mature evaluation technology in competitiveness research is adopted to conduct comprehensive evaluation on the GEC of all countries; evaluation results are also thoroughly interpreted and analyzed, and compared horizontally and vertically. Also under analysis are the comparative advantage and history of each country, the causes for such advantage or disadvantage, and the barriers of enhancing competitiveness. These analyses are not only aiming at evaluation of history and the present, but also the intrinsic factors that influence competitiveness. Judgment and prediction of the trend of competitiveness development is likewise done.

As for application of the evaluation results, focus is put on the integration of theory and practice. Evaluation results are the objective reflections of things and therefore should be used to better guide the development of the things. Of course, evaluation itself is not the purpose, but an instrument; evaluation results are neither simply rankings nor can be more visualized scores to give a better image of GEC. On the one hand, horizontal and vertical comparison of GEC may found out the advantages and disadvantages of all countries, so as to summarize the basic features and trend of development of GEC; thus the key indicators that constrain and influence the GEC of all countries, the weak link and its root, as well as the trends of GEC can all be found out. With these findings, relevant countermeasures can be proposed to help the enhancement of GEC. On the other hand, through GEC evaluation and analysis, it will be good to raise people’s awareness of the importance of environmental protection and ecological economic development; awareness of enhancing GEC will be converted into feasible actions to make new contributions to the global sustainable development.

2 GEC Indicators Selection and Data Source

2.1 Selection of Indicators

Owing to the different understanding of GEC, the designed factor module may be very different, and so are the way to construct the index system and the method to select the indicators; therefore the final evaluation results would be widely divergent. Index system is the core of evaluation and the carrier of evaluation procedures and results; whether or not a complete and objectively applicable index system can be constructed is the key to successful evaluation. First, it is very important to make the process of construction always surround the connotations and definition of GEC. Design of factor module and verification of it are also necessary, because these help to define the scope for selecting indicators and are also the reference for optimization of the index system. Secondly, there must be principles followed during construction of the index system as criteria of screening; only indicators screened via the principles can be included by the system. Finally, the system layer, factor layer and foundation layer are designed for the index system and each indicator is selected with breaking down of the layers and after several rounds of expert discussions and the final complete GEC evaluation index system is confirmed under repeated consideration. The confirmed index system is composed of 1 index, 5 Sub-index, 16 Pillars ad 60 Individual indicators; each of the individual indicators is objective indicator carrying statistical data, which avoids the impact of uncertain and discrete subjective indicators on the impartiality of evaluation results.

2.2 Data Collection, Statistics and Calculation

Data are the basic elements of GEC evaluation; the authenticity of data directly influences the quality of evaluation results; therefore, source of data is of vital importance to evaluation results. Although the United Nations has unified the System of National Accounting (SNA) as a reference to all countries, as the countries have different state system and at different stage of development, there will be distinct differences while doing national economic accounting, especially in terms of scope of statistics, statistical range and statistical time period, which severely influences the comparability of even the same indicator in different countries.

In 1993, the United Nations formally released the System of Integrated Environment and Economic Accounting, which is featured in taking SNA as the basis to build satellite accounting covering various natural resources and environmental ecological fields and which connects the accounting of natural resources and environment with the traditional national accounting. This system added large number of estimation methods about resource consumption and reduction and environmental degradation, accompanied with enormous indicators about resources and environment. But, as the theory about resource environment accounting is not mature, practice in this area shows many problems and weaknesses; consequently, many countries failed to establish a complete accounting system, either with incomplete indicators, or inaccurate.

These problems make the selection of indicators and collection of data for this study more difficult, which actually become a bottleneck of GEC evaluation and research. In order to guarantee objectiveness and impartiality of the data source, here are the principles to be followed during selection of indicator and collection of data: (1) Better use a less number of indicators as possible, trying to select the typical indicators that can reflect the influence on GEC in certain aspect and avoiding excessive influence of the indicators on data collection; (2) Select general indicators, or the universally recognized and frequently used indicators in related researches, avoiding using obscure indicators with unclear definition or ambiguity in meaning; (3) Collect data only from international organization sources such as the UN and World Bank to guarantee the uniform scope of statistics and comparability, statistical yearbook of the countries as the alternative source of missing data. Description of the indicators and source of the data are given in Appendix I. The sources listed in the appendix means the key channel of data collection, mainly the UN, World Bank and International Energy Agency that have provided the majority of data for the countries; but many indicators lack data for certain countries, and these are obtained from the statistical yearbook or government sector official website. As these sources are in great number, details are omitted for convenience.

2.3 Data Extreme Value Analysis

Among the substantial statistical indicator data, it is inevitable to have some “noise” data (maximum or minimum value), i.e. individual datum that shows big difference from the majority of the data. Such phenomena might be the problem of the indicators due to the wide gap between themselves, or the error during the process of data collection and reorganization. Particularly under current circumstance when the resource environment statistics system is far from sound, statistical survey and method of reorganizing the data might both lead to “noise” data. The numerous indicators in GEC evaluation indicator system involve many entirely new areas, and some, including resource and environment areas, do not have well established sound statistical system; actually, some of the statistical methods are still under modification. These are all challenges for the authenticity and objectivity of the GEC indicator data. In addition, the geographical scope of evaluation covers more than 130 countries widely distributed around the globe and the national conditions in each country are varied; it is quite possible to see data error in the process of accounting. The existence of “noise” data is a negative factor for the evaluation of GEC. Especially, the evaluation adopts comprehensive weighting method, under which the comprehensive competitiveness score is obtained from the weighted score of the lower-layer indicators and the bottommost individual indicator scores are obtained from the non-dimensional value of evaluation samples by efficiency coefficient method; in other words, the score of single indicator will affect the total evaluation score through weighting layer by layer. If some indicator carries maximum or minimum value, the scores of the samples calculated according to the non-dimensional formula by efficiency coefficient method will be enormously different and the distribution of evaluation scores turn to be irrational, which all influences the evaluation result. In addition to analysis on the characteristics of each indicator and making judgment, it is also fully necessary to find possible extreme values of the indicator using appropriate quantitative approach and process the extreme values.

The judgment of extreme value is carries out according to the variance of data distribution. Indicator data shows certain distributional characteristics among the samples and the distance between each datum and their average value always follows certain laws and is related to the standard deviation of the sample data. Suppose data are in normal distribution, then 99.97 % of the data will be distributed within the range of 3 standard deviation of the average value, i.e.:

$$ P\left(\Big|\left(x-\overline{x}\right)/\sigma \Big|<3\right)=0.9997 $$
$$ \sigma =\sqrt{{\displaystyle \sum }{\left(x-\overline{x}\right)}^2/\left(n-1\right)} $$
(5.1)

\( \overline{x} \) is the average value of sample data and σ is the standard deviation of sample data. Of course, the actual distribution of indicators would not be strictly in normal state, but according to the Law of Large Numbers, even the indicator data is other state of distribution, such feature also exists. So, if certain sample value of the indicators goes beyond the range of 3 standard deviation of the average value, the value can be judged as the extreme value of the indicator and needs regression to within the range after treatment of re-check and revision.

3 GEC Indicator System Correlation Analysis

In the process of GEC evaluation, setting up the index system is a core step. In order to adequately reflect the different factors that influence environmental competitiveness, the index system becomes huge with enormous indicators and covers substantial contents. The merit of such setting is to avoid insufficiency of information because of too small number of indicator and to reflect multiple aspects of GEC. At the same time, such arrangement can prevent improper influence on the evaluation results caused by extraordinary fluctuation of individual indicator, unless the number of indicator is too small; in this way, the evaluation results will be ore stabilized and rational. But, the problem faced during construction of the comprehensive index system is that the indicators, more or less, shows correlation, or, different indicators containing same information; actually, during the process of evaluation, repetition of information is quite often. If the contents reflected by two indicators are similar or of the same nature, then the indicators contain repeated information; and if both of the indicators are included by the indicator system, the consequence is overlap** of indicator and information redundancy, or even contradiction. During evaluation, this part of information would be calculated doubly, which influences the precision of evaluation results. The indicators in the GEC evaluation index system cover multiple aspects including ecological environment, resource environment, environmental management, environmental carrying and environmental harmony, 5 Sub-index, 16 pillars and 60 individual indicators in total. There has been large amount of information commonly reflected by indicators, particularly those, that are related to economic and social activities, very often showing strong correlation in between. This is also bad for analysis on the driving power of competitiveness. Therefore, a correlation analysis on the indicators should be done first. When obvious correlation is diagnosed, relevant treatment is necessary to remove such correlation.

Indicator correlation analysis is a study of whether there is dependent relation between existing phenomena and discussion of the direction and degree of correlation in specific phenomena having dependent relationship; it is a kind of statistical method to study the correlativity between random variables. By the direction of changing in the two variables, correlativity includes positive correlation, negative correlation and no correlation. (1) Positive Correlation: When one variable increases or decreases, the other variable is also increasing or decreasing and the directions of changing for both variables are the same, which is called positive correlation. (2) Negative Correlation: When one variable increases or decreases, the other variable is also decreasing or increasing and the directions of changing for both variables are opposite, which is called negative correlation. (3) No Correlation: Between two variables, the change in one variable is not related to the change of the other variable; such relationship is also called zero correlation. Of course, such classification is only a simple judgment of the relationship between two variables, which is not precise. A more precise statistical indicator is needed to reflect such relationship between two variables, i.e. using a statistic to reflect the correlation between two variables. According to the type of variable data, different calculation method should be used. GEC indicator system data are continuous variable using scale and dimension of definite proportion and therefore can use “product moment method” to calculate the correlation coefficient, measuring the degree of correlation. This method uses the product of the dispersion of the two variables and the respective mean value, i.e. Pearson’s formula:

$$ {r}_{xy}=\frac{{\displaystyle \sum}\left(x-\overline{x}\right)\left(y-\overline{y}\right)}{\sqrt{{\displaystyle \sum }{\left(x-\overline{x}\right)}^2}\sqrt{{\displaystyle \sum }{\left(y-\overline{y}\right)}^2}} $$
(5.2)

x and y are the two variables to be measured in terms of correlation coefficient. r xy is the coefficient, reflecting the statistic of correlativity between x and y, also called simple correlation coefficient. The sign symbol of r xy determines the positive or negative correlation between x and y and the value of r xy is between −1 and 1. The closer the absolute value is to 1, the higher the correlation between x and y will be; vice versa, the closer the absolute value of ρ xy is to 0, the less obvious the correlation between x and y will be. There is reference standard to judge and test the correlativity, by the test statistic of:

$$ t=\sqrt{\frac{r_{xy}^2\left(n-2\right)}{1-{r}_{xy}^2}}\sim {t}_{\partial /2}\left(n-2\right) $$
(5.3)

Although correlativity only reflects the relevancy between two indicators, in a comprehensive indicator system, the relationship between multiple indicators is complicated, mutually influencing and interrelated. Multiple correlation is right the study of correlativity between one variable and another set variables; it can reflect the correlation of multiple indicators. The philosophy behind this is the same as simple correlation coefficient; the larger the value, the closer relation between the variables. It is generally used in multiple regression analysis and suitable for factor analysis.

Through calculation of the correlation coefficients between each layer of indicators, the summarized results after test of significance are given in Table 5.1.

Table 5.1 Correlation analysis on GEC indicators
Full size table

The indicator correlation statistics show that correlation between the original data of some environmental competitiveness indicators is relatively obvious and that the correlation coefficient between the four pillar and the subordinate individual indicators is relatively larger. More number of correlation coefficient that passes the significance test indicates that many original indicators show higher correlation. But, except that the individual indicators show certain correlation, the correlation between sub-index and between pillars are not high, which means little influence on the calculation of comprehensive evaluation score and the reliability of both scores and rankings of GEC.