Printer Friendly

Research on government's response capability to network opinion on public emergencies: measurement model and empirical analysis.

1. Introduction

At present, our country is experiencing an important period where economy is developing rapidly. The economic structure change in interest's pattern and values highlights the social contradictions, resulting in frequent network emergencies. Via new media, the public have expressed their opinions around the natural disasters, public safety, corruption and other public emergencies. Under the new media environment, diffused and uncontrollable information communication, weakening doorkeeper role and public selfhood agenda-setting, all these factors caused the traditional government guide inapplicable, which challenged the traditional methods and thus brought a great impact on the government response capability on network opinion (Tsai, 2016, Sa et al., 2016).

Facing the high rocketing network public opinion, governments at all levels basically followed the traditional media management philosophy, mainly by means of technology to control network media content and source of information (Shao and Wang, 2016). What's more, government strengthened supervision and controlled censorship on the network public opinion and the Internet information content and finally formed the Internet governance structure basing on the content management (Xu et al. 2013). However, in the process of the government dealing with the network opinion, there are still some problems, for example their negative attitude, improper measures and low efficiency. From the domestic emergency cases, the academia carried on research on the characteristics and evolution of network public opinion, Internet users' network voice and behavior, network communication carrier, Internet space served as public area together with the management path and mode by referring to crisis management and public management theory. Although domestic and foreign scholars analyzed the characteristics, types, process in which government dealt with the network public opinion from the theoretical, practice and macro aspects. Though most of them proposed countermeasures in theory, however, there still lack of structured and quantitative analysis to practice as policy design tools (Su, 2015).

Therefore, to solve these problems, combined with characteristics of public emergencies, it is conductive to carry out effective strategy of government's response to the network public opinion under new situation to improve the management level on network public opinion and virtual social management mode. And it also has important practical significance and theoretical value to establish and improve the comprehensive management of social system. 2

2. Related research

The interactive, open and grass-roots characteristics of public opinions spread of new media pose new challenges for management of public opinion on emergencies. The fixed top-to-bottom management mode on public opinion, when applied to Internet public opinion management, demonstrates some problems like shortage of emergency preparedness, excessive information control, the lack of mainstream media guide, the out-date means of dictation and analysis, and so on. Thus, Chinese government began synthetically using legislative, administrative and technical means and applied network monitoring and analysis of Internet and regulation on public opinion. While classical communication theory is hard to explain the communication practice of networks and their potential impact on the public opinion and so it fails to provide theoretical support for online public opinion guidance and response. Because of these defects, scholars begin to study the public opinion on Internet from management, response, detection, guidance and control from the perspective of government by using the theory of public management and good governance, and by adopting the methods of system dynamics, game theory, differential equations, etc. (Hofmann et al, 2016).

Li and Wang (2014) used a structured description method to analyze the generation of public opinion and its spread dynamic system and then established structure description frame of the Game relationship and the decision behavior, and adopted the evolution game theory to study the Game relationship of the public and the public with the Government; and finally the case has proven the effectiveness of this model.

Yu et al. (2015) proposed the netizens--Internet media--Government model of network public opinion emergency for Government based on system dynamics, with the emergency of the water chemical pollution as empirical object, exploring the regulation of Internet public opinion spread and evolution as well as the effective strategy of emergency management, and he held the view that the improvement of official news transparency, Government crisis processing efforts, and Government response efficiency can control and guide Internet public opinion from diffusion and spread effectively.

Based on capacity views, Chen and Ye (2013) established a system dynamics model of the government response capability on the Internet public opinion by using the theory of system dynamics from four dimensions of Government, netizens, Internet media and traditional media. The simulation results showed that, timely and accurate response of public demands, improving the governments' attention on public opinion and response speed and credibility can effectively enhanced governments' efficiency and effect.

As it can be seen from the above analysis, the government response ways on public opinions have been changing with the change of transmission mode, the shifting of modern society and the change of public opinion expression. Therefore through the case study, this paper tries to define the concept of the government's response capability on network opinion and establish a concept model from the capacity views.

3. Construction of the concept model of the government response capability to network opinion

3.1. Government response capability and its key elements

Many scholars described government response capability to network opinion from emergency management, government functions and other aspects, but has not yet achieved a unified definition. For example, Cao and Huang (2015) thought that emergency management ability means the local governments' response performance with collaborative departments shown in natural disasters and emergencies, fulfillment degree of public safety responsibility and public management functions in risk society. It reflects the subjective conditions and the comprehensive quality of the government emergency management. Du (2013) hold that "Government response capability" is the comprehensive ability when facing with a challenge from network opinion supervision, all levels of government organs and their staff should effectively mobilize the most social resources to carry out prevention, detection, analysis, treatment and rehabilitation of the network public opinion crisis in a relatively short period of time, real properly collect and disclose public information, scientifically and effectively construct and execute system, as well as communicate and coordinate well.

In order to reflect the consistency and prevent confusion, this paper deals with the concept of government response capability and key elements of network opinion through the relevant literature review, the use of content analysis method to refine the connotation and key factors. Table 1 lists some concepts of government response capability to network opinion.

From Table 1, this paper holds, government response capability to network opinion is the ability to deal with network opinion in the Internet era. It is a creative ability for governments to keep its energy, adapt to always changing online environment, insist on people oriented, and realize the value of the whole society. The response capacity of network opinion government consists of many factors such as timely response, open information and emergency coordination.

3.2. Index system construction

In the process of government's response to network opinion, if there are no scientific and reasonable indicators to measure government's response capability to network opinion, and no dynamic reflections of government's response capability in different stages of network opinion evolution, not only the response to network opinion cannot achieve the corresponding effect, but also the government will lose enthusiasm due to the difficulties of responding. Due to this reason, based on the concept and key elements of the government's response capability to network opinion and combined with the case summary, this paper holds that when disposing such network opinion arising from public emergencies, government departments at all levels should be concerned about the emotions and aspirations of public Internet users, publish relevant information content of emergencies efficiently and effectively to make information open and transparent, as well as enhance their credibility by coordinating with related departments and responding in accordance with specified procedure. By integrating common points of different perspectives, this paper considers authority, exactitude, diaphaneity, trust, timeliness, standardization, coordination as the key elements of the effectiveness of government's response capability to network opinion on public emergencies, namely, seven basic dimensions.

Embarking on the above seven basic dimensions and based on plenty of literature reference and analysis of typical public emergencies, this paper establishes an index system to measure government's response capability which is composed of 7 first level indexes and 21 second level indexes (as shown in table 2).

4. Government's response capability measurement

The initially constructed index system (Table 2) consists of some major information of response to network opinions on public emergencies. However, there is often a strong correlation and repeatability among indexes and a wide range of indexes will affect the efficiency and exactitude of the calculation of indexes. Therefore, this paper uses rough set to simplify the original index system, and improve the efficiency of response capability measurement.

Rough set is a mathematical tool to describe the incompleteness and uncertainty, which can analyze imprecise, inconsistent, incomplete and all kinds of imperfect information efficiently and can carry out the identification of the equivalence relation, attribute reduction and implicit knowledge discovery under the premise of retaining information (Pawlak, 1982). Compared with analytic hierarchy process, fuzzy comprehensive evaluation, neural network, etc., rough set doesn't need any prior information such as fuzzy membership function, basic probability assignment and relevant statistical probability distribution other than a decision table. Under the premise of retaining the key categories of knowledge, it can reduce attributes, simplify decision table and find minimum expression of classification knowledge by identifying and evaluating dependency relations, so as to overcome the shortcomings of traditional methods, such as strong subjectivity, poor testability and so on(Hu et al. 2001; Hu and Yi, 2016).

Based on the characteristics of rough set above and combined with the uncertainty of evaluation of response capability to network opinion, this paper simplifies the initial indexes of government's response capability to network opinion on public emergencies, screens and refines minimum index set for classification knowledge of government's response capability to network opinion on public emergencies by referring to attribute reduction algorithm of rough set for reference, so as to improve the calculation efficiency of evaluation and prediction.

4.1. Building decision table system

Based on network opinion monitoring data from People's Daily and public opinion database of Communication University of China, according to rough set theory, the measurement of response capability to network opinion on public emergencies can be abstracted as 4 tuples:S=<U, R, V, f>. Decision table S is an evaluation system of government's response capability to network opinion on public emergencies (as shown in Table 3). U={[e.sub.i]|i=1, 2, ... m} is discourse domain, which is composed of emergencies [e.sub.i]. Ris an attribute set, with C={[I.sub.i]|i=1, 2, ... n} being condition attributes, [I.sub.i] being the ith index and D={d} being decision attributes. Vis the domain value of attribute indexes. f: UxR [right arrow] V specifies every attribute value of [e.sub.i] in the U.

4.2. Attribute discretization

Before doing attribute reduction via equivalent class relation in rough set theory, attributes must be discretized into a finite number of semantic symbols. Commonly used discretization methods include the equal distance method, the equal frequency method, etc. (Zhang et al. 2005). Equal distance (equal value domains of continuous attributes) and equal frequency (each discrete interval with the same object) are common unsupervised discretization algorithms. Though easy to implement, they can't set the interval boundary on the most appropriate breakpoint because of ignoring the sample distribution information, therefore data discretization performance are influenced (Wang 2007).

For this reason, this paper uses Fuzzy C-means clustering analysis method to discretize attribute value. On the basis of maintaining the distribution characteristics of the original sample, it aims to divide similar samples into a class and distinguish dissimilar samples. It provides discrete value of data, center of each fuzzy region and degree of membership of each attribute values in cluster center and takes the class number of each column attribute value as the characteristic values after the discretization, which are often described as "1", "2", " 3". Based on the algorithm, by applying the fuzzy logic toolbox in matlab to complete the operation of the cluster. Index system after discretization is as shown in Table 3.

4-3-Index selection based on the diseernibility matrix

Definitions Decision table S=<U,R, V, f>, discourse domain U={[e.sub.i]|i=i, 2, ...m}, attribute set R=C[union]D, condition attribute C={[a.sub.i]|i = 1, 2, ...n}, decision attribute D={d}, [a.sub.i]([e.sub.j]) is the value of the object [e.sub.j] in the attribute [a.sub.i], [C.sub.D](i, j) represents the element of row i, column j in an identifiable matrix, and diseernibility matrix CD:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (1)

After the diseernibility matrix is obtained, decision table can be simplified. Specific steps are as follows:

Step 1: Based on given discourse domain, construct the corresponding decision table S.

Step 2:Calculate the discernibility matrix [C.sub.D] of decision table S.

Step 3: For every element [C.sub.D](i, j) of a non-empty set in discernibility matrix, establish corresponding expression of disjunctive logic L(i, j)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

Step 4: Put all the disjunction logic expressions L(i, j)in conjunction operation and get the conjunctive normal form L

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

Step 5:Transform conjunctive normal form L to the conjunctive normal formL',

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

Step 6:Output the result of attribute reduction. Attributes in every conjunctive items form a condition attribute set after reduction.

Based on the concept of discernibility matrix, according to formula (1) to (4), an index system reduction is done. Here we take authority, a first level index, and its second level index I11, I12, I13 as an example(as shown in Table 4.). And then, we construct discernibility matrix [C.sub.D] according to formula (1) (as shown in Table 5). According to Table 5, we get expression of disjunctive logic:

[L.sub.1,2] = [I.sub.11] [disjunction] [I.sub.12], [L.sub.1,3] = [I.sub.11] [disjunction] [I.sub.12], [L.sub.1,4] = [I.sub.11] [disjunction] [I.sub.12] [disjunction] [I.sub.13], ... [L.sub.9,10] = [I.sub.12] [disjunction] [I.sub.13]

Get the disjunctive normal forms [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], using formula (2) - (4)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The three main implications of the formula are all reductions, and attribute reduction {I11, I12} is obtained. Combining the result of case analysis, this paper considers I11, I12, i.e. Response hierarchy, Affair guidance as the second level index of authority.

By repeating the above steps, similarly, we can get attribute reduction of second level index of the six dimensions: exactitude, coordination, diaphaneity, timeliness, standardization and trust. It can be found that when deleting any index [I.sub.j] [member of] {I22, I33, I41, I43, I61, I63, I73} from network public relations (I22), media coverage(I33), service quality (I41), integrity degree (I43), regulatory framework (I61), resource reservation (I63), professional institutions (I73), [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. Therefore index [I.sub.j] isn't necessary to decision attribute d .Network public relations (I22), media coverage (I33), service quality (I41), integrity degree (I43), regulatory framework (I61), resource reservation (I63) and professional institutions (I73) can be deleted from table and we can get an index system after reduction, which is made up of 7 first level indexes and 12 second level indexes, as shown in Table 6. 5

5. Empirical analysis

To test the effectiveness of government's response to network public opinion, this paper takes the rankings of 2011-2012 local government's response capability to network opinion released by People's Daily (www.people.com.cn). Then this paper selects 50 network public emergencies from 2011 to 2012 to design corresponding questionnaires and distribute them to experts for assessment.

Lasting for 3 months, this questionnaire survey combines paper version questionnaires with electronic version questionnaires. Respondents cover domestic municipal governments and some middle-level cadres of state owned enterprises.400 questionnaires are sent out, including 200 electronic version questionnaires and 200 paper version questionnaires. After 3 rounds of distribution, 273 valid questionnaires are received and the response rate is 68.25%. We use SPSS for data preprocessing, which can analyze the validity and reliability of each measurement variable and can get a factor loading matrix after rotation. As Table 7, the design of the questionnaire is reasonable and the questions can basically reflect the classification of the first level index. Besides, the validity of the second level index to first level index is reflected on the variance contribution rate of the first principal component. The higher the variance contribution rate is, the better the construct validity of the questionnaire is. A general criteria for variance contribution is 40%. The reliability of each variable can show the reliability of the questionnaire itself. From Table 7, we can find that the value of the reliability coefficient of each variable is between 0.7 and 0.9.

According to the established index system above, we construct a multiple linear regression model between Network opinion incidents' effect and 13 second level indexes. As shown below:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

In the formula above, Y represents the effect of government's response, which values 1 when positive while 0 when negative. Elements between [[alpha].sub.0] and [[alpha].sub.13] are coefficients of the explanatory variables. X1 to X13 represent response hierarchy, affair guidance, professional level, media literacy, information disclosure, press conference, actions by regulation, speed of response, response attitude, response frequency, emergency plan, information sharing and department linkage respectively. All variables are independent categorical variables. [epsilon] is error term, assuming that it is not related to the other independent variables.

During the regression, the 13 second level indexes of the index system after reduction and the effect of government's response are regressed and the result is shown in Table 8.

From the regression result in the Table 8 above, it is shown that the value of Multiple R-squared is 0.5941, which is close to 1. The value of adjusted R-squared of the sample after being adjusted by sample size and degree of freedom is 0.515, which means the reference value of correlation equation is high. The value of F-statistic is 3.318, showing that the possibility of existence of null hypothesis is true and p-value<0.0l. Generally, the index system is significant. According to the regression results, this paper comes to the following conclusions:[R] From the significance of the index, the regression result shows that 11 indexes are significant including response hierarchy, affair guidance, Information disclosure, press conference, etc. It shows that the effect of government's response is mainly affected by these factors. While the effects of actions by regulation and emergency plan on government's response capability are not significant, thus these two indexes, actions by regulation and emergency plan (X7, X11) are deleted. Meanwhile, these two second level indexes, actions by regulation and emergency plan belong to two dimensions, trust and standardization respectively. Therefore after general regression, the number of dimensions of index system of the ability to response to network opinion on public emergencies decrease from 7 to 5. The number of second level index decrease from 13 to 11 (as shown in Table 9). (ii) According to the results of regression, the significant indexes which have more impact on the effect of government's response are press conference, speed of response, response hierarchy, information sharing, affair guidance. It shows that if we want to improve the effect of government's response to network opinion on important incidents, the most important thing is to release information in time and to strengthen the guidance of the situation through sector linkage.

According to Table 9, the five dimensions of government's response capability to network opinion are timeliness, diaphaneity, authority, exactitude and coordination. Among these, timeliness is measured by the speed of response, response attitude and response frequency. Diaphaneity is measured by information disclosure and press conference, which means that after public emergencies, government should release related information to public, meanwhile it should decide whether or not to answer the public's concern by holding a press conference according to the order of priority. Authority is measured by response hierarchy and affair guidance, which means that relevant departments of government should respond and give positive guidance from the form of agenda after public emergency happens. Exactitude is measured by Professional level and Media literacy, which means that response personnel from government should be on the professional level and can give a keen judge on the development of the situation. Coordination is measured by information sharing and department linkage, which means every departments in government should keep the information exchange, be united in concert and respond jointly in order to avoid the phenomenon of different lines and buck-passing.

6. Conclusions

Facing the high rocketing network opinion, the government's response still have some problems at present like negative attitude, inappropriate measures, low efficiency and so on. Domestic and foreign scholars only analyze the characteristic and patterns of government's response to network opinion from theoretical level, practical level and macro level, which lack structural analysis and quantitative analysis and are inefficient in operation.

Considering the above drawbacks, we summarize key elements of government's response capability to network opinion and constructs an index system of ability to response including 7 first level indexes and 21 second level indexes. In order to improve the computational efficiency of evaluation index, based on attribute reduction algorithm of rough set theory, an index decision table is built, which discretizes index data, simplifies the original index system and simplifies the number of second level indexes from 21 to 13. Finally, based on data from questionnaires of 50 typical cases from 2011 to 2012, an empirical analysis is carried out by the regression model constructed. Finally this research establishes an index system of government's response capability to Network opinion on public emergencies which is made up of 5 first level indexes including authority, exactitude, diaphaneity, timeliness and coordination and 11 second level indexes including response hierarchy, affair guidance, professional level and media literacy etc. The empirical result shows that the index system is feasible and effective. In the future the research should delve deeper into how to combine the index system with the evolution process of public emergencies so as to analyze government's response capability to network opinion dynamically, which will be of practical guiding significance.

Recebido/Submission: 15/07/2016

Aceitacao/Acceptance: 10/10/2016

Acknowledgments

The authors would like to thank the anonymous reviewers. The research supported by Key projects of the National Natural Social Science Foundation of China (Grant No. 14AZD045), the Ministry of education of Humanities and Social Science (Grant No. 13YJA630089).

References

Cao, H. M., Huang, Y.N. (2015). A Research on the Evaluation Index System of Local Government Emergency Management Capabilities. Journal of Guangzhou University (Social Science Edition), 14(12), 60-66.

Chen, F. J., Ye, M. (2013). Improving Government's Ability in Responding to Internet Public Opinion: A System Dynamics Perspective, Journal of Intelligence, 32(11), 121-127. Doi: 10.3969/j.issn.1002-1965.2013.11.023

Du, X.Z. (2013). Study on Government Ability to Respond to Network Opinion and Strategies for Improvement. Journal of Zhengzhou University (Philosophy and Social Sciences Edition), 46(5), 20-23.

Han, Z. M. (2007). An Analysis on the Uncertainty of Government Management Ability in Public Crisis. Hubei Social Sciences, (3), 15-18. Doi: 10.3969/j.issn.1003-8477.2007.03.004

Hofmann, J., Katzenbach, C., Gollatz, K. (2016). Between Coordination and Regulation: Finding the Governance in Internet Governance. New Media and Society, 1461444816639975. Doi:10.1177/1461444816639975

Hu, K. Y., Lu, Y. C., Shi, C. Y. (2001). Advances in Rough Set Theory and its Appliations. Journal of Tsinghua University, 41(1), 64-68. Doi: 10.3321/j.issn:1000-0054.2001.01.017

Hu, W., Yi, M. (2016). Network Reliability Study based on Analytic Hierarchy Process(AHP). RISTI-Revista Iberica de Sistemas e Tecnologias de Informacao, 18, 165-175.

Hu, Y. Z., Huang, Z. B. (2010). Enhancing the government responsiveness in the perspective of public participation. Chongqing Administration, 12(3), 44-46.

Li, Y. J., Wang, Z. Y. (2014). Evolutionary Game and Prpagation Mechanism of Public Opinion in Emergencey, Chinese Journal of Management Science, 22(11), 87-96. Doi: 10.16381/j.cnki.issn1003-207x.2014.11.001

Niu, C. H., Ren, C. J., Bai, J. X. (2003). The Measurement of the Government's Emergency-dealing Ability in Unexpected Public Accidents.Journal of Taiyuan University of Technology (Social Sciences Edition), (4), 21-25. Doi: 10.3969/j. issn.1009-5837.2003.04.007

Pawlak, Z. (1982). Rough set. International Journal of Information and Computer Science, (11), 341-356. Doi: 10.1007/BF01001956

Qi, M. S., Li, Y. Y. (2008). Analysis on Lack of Capacity of the Government Crisis Management and Its Enhancing Path in the Period of Social Transformation. Journal of Tianjin Administration Institute, 10(6), 26-32. Doi: 10.3969/j. issn.1008-7168.2008.06.006

Sa, F., Rocha, A., & Cota, M. P. (2016). Potential dimensions for a local e-Government services quality model. Telematics and Informatics, 33(2), 270-276. Doi: 10.1016/j. tele.2015.08.005

Shao, P., Wang, Y. (2016). How Does Social Media Change Chinese Political Culture? The formation of Fragmentized Public Sphere. Telematics and Informatics. Doi: 10.1016/j.tele.2016.05.018

Su, S. (2015). Influence of Network Public Opinion on China's Diplomatic Decision-making. International Journal of Human Rights and Constitutional Studies, 3(1), 19-30. Doi: 10.1504/IJHRCS.2015.069236

Tsai, W. H. (2016). How 'Networked Authoritarianism' Was Operationalized in China: Methods and Procedures of Public Opinion Control. Journal of Contemporary China, 1-14. Doi: 10.1080/10670564.2016.1160506

Wang, D., Wu, M. D. (2007). Rough Fuzzy C-Means Algorithm and Its Application to Image Clustering.Journalof National Univeristy of Defense Technology, 29(2), 76-80. Doi: 10.3969/j.issn.1001-2486.2007.02.017

Wang, F. Y. (2004). On the Ways of Increase of Government Responsiveness. Guangxi Social Sciences, (4), 22-24. Doi: 10.3969/j.issn.1004-6917.2004.04.007

Xie, Z. Z. (2009). On the Basic Way of Improving the Respondent Power of Government. Journal of Inner Mongolia Agricultural University (Social Science Edition), 11(2), 257-259. Doi: 10.3969/j.issn.1009-4458.2009.02.100

Xu, X.L, Chen, Q., Zeng, R.X. (2013). Several Important Issues in China Virtual Society Governance Research, Chinese Public Administration, 11, 7-11. Doi: 10.3782/j. issn.1006-0863.2013.11.01

Yu, L. A., Li, L., Wu, J. Q., Tang, L. (2015). Emergency Policy Exploration for Network Public Opinion Crisis in Water Pollution Accident by Hazardous Chemicals Leakage based on Systematic Dynamics, Systems Engineering-Theory and Practice, 35(10), 2687-2697.

Zhang, G. X., Jin, W. D., Hu, L. Z. (2005). Generalized Discretization of Continuous Attributes in Rough Set Theory. Control and ONTROL AND Decision, 20(4), 372-376. Doi: 10.3321/j.issn:1001-0920.2005.04.003

Wei Zhang (1), Weidong Yu (1), Peng Cui (2) *, Jianbo Wen (3), Tianmei Wang (1)

weizhang@cufe.edu.cn, kddzw@163.com, cuipeng800209@163.com, krisuvm@126.com, wangtm2008@gmail.com

(1) School of information, Central University of Finance and Economics, 39 South College Road, Haidian District, Beijing 100081, China,

(2) Academic Affairs Division, Central University of Finance and Economics, 39 South College Road, Haidian District, Beijing 100081, China,

(3) School of Foreign Studies, Central University of Finance and Economics, 39 South College Road, Haidian District, Beijing 100081, China.
Table 1--Critical factor of government's response capability

References             Definition

Xie (2009)             It is the ability of the government to provide
                       effective public goods and service in time to
                       the public according to demand.

Hu, and Huang (2010)   It is the ability of the government to meet
                       the proper requirements of society by public
                       power and social resources within its own
                       duty.

Wang (2004)            It is the ability of the county government to
                       react to the changes of society.

Niu, et al. (2003)     It is the ability of the government to deal
                       with emergencies, to react without delay aimed
                       at minimizing the loss while facing public
                       emergency events.

Cao, and Huang (2015)  It is behavior of the local government along
                       with its synergetic departments in reacting
                       and dealing with natural calamities and public
                       emergencies and the level of performing in
                       public secure duty and public management
                       function in a risk society. It reflects
                       subjective conditions and comprehensive
                       diathesis of government emergency management.

Qi, and Li (2008)      It is the management of details in the
                       government in dealing with emergencies,
                       including the predictions, estimates, control
                       and others after the emergencies.

Han (2007)             The essence of government emergency management
                       is the function that the government exerted in
                       resource action and spirits capacity in public
                       emergencies.

Du (2013)              Facing the challenge of network supervision of
                       public opinion, the ability to mobilize the
                       most social resources effectively in a
                       relatively short period of time to carry out
                       prevention, detection, analysis, treatment and
                       rehabilitation of the network public opinion
                       crisis. Also to properly collect and
                       appropriate disclosure of public information,
                       ability of executive ability construction of
                       science the effective system, comprehensive
                       ability and good social communication skills
                       etc..

References             Critical Factor

Xie (2009)             Timeliness; the level of service

Hu, and Huang (2010)   Reaction; response

Wang (2004)            Response

Niu, et al. (2003)     Timeliness; comprehensive management; the
                       level of service

Cao, and Huang (2015)  Coordination; comprehensive diathesis

Qi, and Li (2008)      Prediction; supervision; control

Han (2007)             Openess; government image

Du (2013)              Timeliness; information disclosure;
                       communication and coordination

Table 2--Index System of response capability to network opinion on
public emergency (preliminary)

Target Layer    First Level Index      Second Level Index

Government's    Authority(I1)          Response hierarchy(I11)
capability to                          Affair Guidance (I12)
response to                            Official accountability (I13)
network         Exactitude(I2)         Professional level (I21)
opinion                                Network public relations(I22)
                                       Media literacy(I23)
                Diaphaneity (I3)       Information disclosure(I31)
                                       Press conference(I32)
                                       Media coverage(I33)
                Trust (I4)             Service quality(I41)
                                       Actions by regulation(I42)
                                       Integrity degree (I43)
                Timeliness (I5)        Speed of response(I51)
                                       Response attitude(I52)
                                       Response frequency(I53)
                Standardization (I6)   Regulatory framework(I61)
                                       Emergency plan(I62)
                                       Resource reservation(I63)
                Coordination(I7)       Information sharing(I71)
                                       Department linkage(I72)
                                       Professional institutions(I73)

Table 3--Index decision table after discretization

      e1   e2   e3   e4   e5   e6   e7   e8   e9   e10

I11   3    1    1    1    3    3    2    2    2    2
I12   3    2    1    1    3    2    2    1    2    1
I13   3    3    3    1    3    3    2    2    1    2
I21   3    2    1    2    1    2    2    2    2    3
I22   1    1    2    2    2    1    1    2    2    2
I23   2    1    1    1    3    2    2    2    2    3
I31   2    2    2    3    2    2    2    3    1    1
I32   1    2    1    3    1    1    2    3    1    1
I33   1    2    2    3    1    1    2    3    1    1
I41   1    2    2    3    2    1    2    1    1    1
I42   2    1    3    1    3    1    2    3    2    1
I43   1    2    2    3    2    1    2    3    1    1
I51   1    1    2    3    2    1    1    3    1    1
I52   1    1    2    3    2    1    1    3    1    1
I53   1    1    2    3    2    1    1    3    1    1
I61   1    2    2    3    2    1    2    3    1    1
I62   1    2    2    3    2    1    2    1    1    1
I63   1    2    2    3    2    1    2    3    1    1
I71   1    2    1    3    1    1    2    3    1    1
I72   1    1    1    3    1    1    2    3    1    1
I73   1    1    2    1    2    1    1    1    1    1
d     1    4    4    4    0    0    3    3    3    2

Table 4--Authority's first level index decision table

e     I11   I12   113   d

e1    3     3     3     1
e2    1     2     3     4
e3    1     1     3     4
e4    1     1     1     4
e5    3     3     3     0
e6    3     2     3     0
e7    2     2     2     3
e8    2     1     2     3
e9    2     2     1     3
e10   2     1     2     2

Table 5--Discernibility matrix [C.sub.D] (under the first level
index: authority)

0   I11I12   I11I12   I11I12I3   0           I12      I11I12I13
    0        0        0          I11I12      I11      I11I13
             0        0          I11I12      I11I12   I11I12I13
                      0          I11I12I13   I11I12   I11I12I13
                                 0           0        I11I12I13
                                             0        I11I13
                                                      0

I11I12I13   I11I12I13   I11I12I13
I11I12I13   I11I12I13   I11I12I3
I11I13      I11I12I13   I11I13
I11I13      I11I12      I11I13
I11I12I13   I11I12I13   I11I12I13
I11I12I13   I11I13      I11I12I13
0           0           I12
0           0           0
            0           I12I13
                        0

Table 6--Index system of government's response capability to network
public opinions (after reduction)

Target Layer   First level index      Second level index

               Authority (I1)         Response hierarchy (I11)
Index System                          Affair guidance (I12)
               Exactitude (I2)        Professional level(I21)
                                      Media literacy (I22)
               Diaphaneity(I3)        Information disclosure(I31)
                                      Press conference (I32)
               Trust (I4)             Actions by regulation (I42)
                                      Speed of response (I51)
               Timeliness (I5)        Response attitude (I52)
                                      Response frequency (I53)
               Standardization (I6)   Emergency plan (I62)
               Coordination (I7)      Information sharing (I71)
                                      Department linkage (I72)

Table 7--Validity and reliability analysis of factor loading matrix
after rotation

Variable   Factor1   Faetor2   Faetor3   Faetor4   Factors   Faetor6

X1         .710      .113      .261      .176      .101      .152
X2         .772      .030      .142      .198      .166      .090
X3         .130      .763      .219      .157      .117      .265
X4         .205      .814      .246      .039      .285      .219
X5         .071      .213      .720      .155      .129      .202
X6         .214      .109      .727      .131      .046      .029
X7         .114      .203      .119      .801      .170      .220
X8         .173      .037      .053      .283      .716      .102
X9         .106      .243      .032      .154      .750      .013
X10        .162      .132      .253      .098      .801      .115
X11        .031      .129      .227      .150      .209      .710
X12        .118      .292      .195      .071      .085      .294
X13        .220      .216      .171      .351      .047      .235

Variable   Factory   Cumulative            Cronbach's
                     contribution          [alpha]
                     rate of the first
                     principal component
                     variance

X1         .230      60.170                0.70
X2         .185
X3         .194      67.021                0.75
X4         .125
X5         .175      68.283                0.76
X6         .173
X7         .189      63.589                0.72
X8         .133
X9         .119      71.301                0.79
X10        .177
X11        .139      78.714                0.82
X12        .740      65.385                0.74
X13        .703

Table 8--Regression analysis results

              Estimate   Std. Error   t value   Pr(> [absolute
                                                value of (t)])

(Intercept)   2.01524    1.10231      1.828     0.0763
X1            -1.04349   0.37162      -2.808    0.0082 **
X2            1.06874    0.51787      2.064     0.04674.
X3            0.58302    0.40479      2.44      0.01189 *
X4            -0.21024   0.40082      -2.525    0.0146.
X5            -1.27056   0.49299      -2.577    0.0447.
X6            1.90165    0.65962      2.883     0.00678 **
X7            -0.544     0.4654       -0.169    0.24951
X8            1.62873    0.47995      3.394     0.00177 **
X9            0.47167    0.35974      2.311     0.0816.
X10           -0.64745   1.32712      -2.488    0.01328 *
X11           0.2436     0.3402       0.716     0.47827
X12           1.16292    0.51108      2.275     0.0113 *
X13           -0.36445   0.71294      -2.511    0.01361 *

Signif. codes:0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1;
Multiple R-squared: 0.5941, ; Adjusted R-squared:0.515; F-
statistic:3.318 on 15 and 34 DF; p-value:0.001843

Table 9--Index system of government's capability to response to
network opinion on public emergencies (final)

Target Layer   First level index   Second level index

Index System   Authority (I1)      Response hierarchy I11
                                   Affair guidance I12
               Exactitude (I2)     Professional level I21
                                   Media literacy I22
               Diaphaneity (I3)    Information disclosure I31
                                   Press conference I32
                                   Speed of response I51
               Timeliness (I5)     Response attitude I52
                                   Response frequency I53
               Coordination (I7)   Information sharing I71
                                   Department linkage I72
COPYRIGHT 2016 AISTI (Iberian Association for Information Systems and Technologies)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Zhang, Wei; Yu, Weidong; Cui, Peng; Wen, Jianbo; Wang, Tianmei
Publication:RISTI (Revista Iberica de Sistemas e Tecnologias de Informacao)
Date:Nov 15, 2016
Words:5914
Previous Article:Survey on e-commerce development in areas inhabited by ethnic minorities of Qujing city, Yunnan province, based on computer data analysis.
Next Article:A novel Bayesian algorithm for evaluation lifetime performance of exponential model.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters