Assessment of online public opinions on large infrastructure projects: A case study of the Three Gorges Project in China

https://doi.org/10.1016/j.eiar.2016.06.004Get rights and content

Highlights

  • We developed a framework to assess online public opinion on large infrastructure projects with environmental impacts.

  • Indicators were built to assess post intensity, sentiment polarity and major topics of the public opinion.

  • We took the Three Gorges Project (TGP) as an example to demonstrate the effectiveness proposed framework.

  • We revealed spatial-temporal patterns of post intensity and sentiment polarity on the TGP.

  • We drew implications for a more in-depth understanding of the public opinion on large infrastructure projects.

Abstract

Public opinion becomes increasingly salient in the ex post evaluation stage of large infrastructure projects which have significant impacts to the environment and the society. However, traditional survey methods are inefficient in collection and assessment of the public opinion due to its large quantity and diversity. Recently, Social media platforms provide a rich data source for monitoring and assessing the public opinion on controversial infrastructure projects. This paper proposes an assessment framework to transform unstructured online public opinions on large infrastructure projects into sentimental and topical indicators for enhancing practices of ex post evaluation and public participation. The framework uses web crawlers to collect online comments related to a large infrastructure project and employs two natural language processing technologies, including sentiment analysis and topic modeling, with spatio-temporal analysis, to transform these comments into indicators for assessing online public opinion on the project. Based on the framework, we investigate the online public opinion of the Three Gorges Project on China's largest microblogging site, namely, Weibo. Assessment results present spatial-temporal distributions of post intensity and sentiment polarity, reveals major topics with different sentiments and summarizes managerial implications, for ex post evaluation of the world's largest hydropower project. The proposed assessment framework is expected to be widely applied as a methodological strategy to assess public opinion in the ex post evaluation stage of large infrastructure projects.

Introduction

The public opinion is crucial for decision making of large infrastructure projects and the public satisfaction is also a key indicator for ex post evaluation of large infrastructure projects (Chan and Chan, 2004, Ji and Hong, 2016, Stoutenborough and Vedlitz, 2014). Successful implementation of large infrastructure projects depends not only on the financial and environmental feasibility, but mainly on the support from the public, who ultimately pays for and will be influenced by these projects (Friedler et al., 2006). Thus, the assessment of public opinion should be included in the delivery framework of large infrastructure projects (Ng et al., 2013). However, the complexity of large infrastructure projects and diverse interests of the public stakeholders make the assessment of public opinion very difficult.

Large infrastructure projects require more investment, consume more public resources and have more profound impacts on local even national economy, society and environment than general projects (Zeng et al., 2015). Hence, they usually have a large number of public stakeholders with a wide range and complicated levels. According to the impacts from projects, public stakeholders can be divided into two categories, including local community and general public. Local communities are directly influenced by large infrastructure projects. On the one hand, local people can benefit from large infrastructure projects, as these projects provide many jobs and public services for them. But on the other hand, if any environmental pollution or safety accident occurs, local people would be the direct victims (Zeng et al., 2015). With these impacts, local people have the motivation to organize substantial opposition activities against the construction or operation of large infrastructure projects if their interests are harmed (Liu et al., 2016). As a result, the acceptance of local people is important to the success of a project. Interests and views of local people on large infrastructure projects are included in current project delivery framework and are fully discussed in the literature (Frynas, 2005, Jobert et al., 2007, Jones and Eiser, 2010, Musall and Kuik, 2011, Wüstenhagen et al., 2007).

The general public is not directly or just slightly influenced by large infrastructure projects, but they are also concerned with the economic and environmental performance of these projects. Comparing with local people, views and comments from the general public would be more objective and unbiased (Zeng et al., 2015). Hence, the public opinion on a large infrastructure project can reflect a more holistic evaluation on that project. In addition, discussion of large infrastructure projects is an important activity in the political life of the public (O'Faircheallaigh, 2010). With the progress on democracy, understanding the public opinion will be more and more critical to policy making in relation to large infrastructure projects. However, in previous practices, public opinions were often overlooked in the delivery framework of large infrastructure projects (Akintoye et al., 2003, Ng et al., 2013). A major reason is that public opinions are difficult to collect and analyze (Ji and Hong, 2016).

Social scientists usually use questionnaire surveys to measure public opinions on conflictive large infrastructure projects (Cihar and Stankova, 2006, Hu et al., 2014). However, traditional survey methods are known as costly, time-consuming and invasive (Kirilenko and Stepchenkova, 2014, Kirilenko et al., 2015). The booming popularity of social media platforms, such as Facebook, Twitter and Weibo (the biggest microblogging site in China), highlights a new opportunity to collect and assess public opinions. User-generated contents in these social media platforms provide massive amounts of data on various topics of social importance. These data make social media as a valuable resource for assessing public opinions on large infrastructure projects, which are controversial and have significant impacts to the environment and society.

This study proposes an assessment framework, which combines two natural language processing technologies, including sentiment analysis and topic modeling, to transform unstructured social media comments on large infrastructure projects into sentimental and topical indicators for enhancing practices of ex post evaluation and public participation. Recent studies regard the interactions among social media users as a large and distributed network of sensors that react to external events by reporting individual comments (Kirilenko and Stepchenkova, 2014, Kirilenko et al., 2015). Using social media content to investigate the public opinion is a non-invasive research method avoiding the complications of interactions with human subjects and has a simple data collection process and a high sampling rate (near 100%). With the massive social media contents, many scholars implement comprehensive analysis of public opinions on natural and social phenomena (Ashley and Tuten, 2015, Gainsbury et al., 2016, Kirilenko and Stepchenkova, 2014, Kirilenko et al., 2015, Quintelier and Theocharis, 2013, Sakaki et al., 2013, van de Belt et al., 2015, Williams, 2013). These applications highlight the potential of using social media data to investigate public opinion on controversial issues, including the construction or operation of large infrastructure projects.

The utility of social media data for public opinion research is augmented by recent data mining technologies, especially sentiment analysis (Pang and Lee, 2008) and topic modeling (Blei et al., 2003). Public comments on social media sites are mainly textual data, which are high-dimensional. Sentiment analysis can map these textual data into an emotional dimension. In other words, whether a comment expresses a negative or positive emotion can be determined, providing an indicator to measure public satisfaction. Topic modeling can uncover the intellectual structure of a large collection of documents. Hence, major topics of the public opinion on a certain issue can be extracted by topic modeling. In the ex post evaluation stage of a large infrastructure project, the government and the management team are also exactly interested in the overall distribution of public sentiment (positive or negative) and the major topics of the public opinion on that project.

Section snippets

Ex post evaluation of large infrastructure projects

By definition, ex post means after the completion, as opposed to ex ante, meaning beforehand (Olsson et al., 2010). Ex post evaluation is a highly valuable tool for determining not only how successful a large infrastructure project may have been after the completion, but rather the long-term performance of the outcomes for the economy, society and the environment (Geurs and van Wee, 2006). Different from final project evaluations, which are performed at the time of a project's completion to

Methods

Based on the literature review, the assessment of online public opinion on a large infrastructure project should be performed from three dimensions, including post intensity reflecting prevalence of discussion on the project, sentiment distribution reflecting public satisfaction with the project and topics revealing the detailed concerns of the public on the project. Post intensity can be simply calculated by basic statistics. Sentiment distribution is generated by sentiment analysis methods

The conflictive Three Gorges Project

TGP is the world's largest hydropower project and is also one of the most high-profile infrastructure projects in China. The construction of the project started in 1994 and completed in 2009 with a total dynamic investment of 248.5 billion Chinese Yuan. After completion, the 181-meter concrete gravity dam raises the normal pool level of the Three Gorges Reservoir (TGR) to 175 m above the sea level, creating a 660-kilometer-long and one-kilometer-wide lake along the Yangtze River (Jiang et al.,

Implications for the TGP

Three Gorges Project is one of the most controversial infrastructure projects of significance to China in many aspects, including politics, economy, national safety and environment. Therefore, it is important to assess the public opinion about the TGP.

The search result demonstrates that the TGP is a constantly attractive topic on China's online public sphere. Weibo users are keen on expressing their opinions about the TGP and averagely produce over 200 messages per day. It further validates

Conclusions

Large infrastructure projects are facing more and more challenges from their environmental and social impacts. These challenges could be tougher with a poor performance on public participation management and the public satisfaction with these impacts should be comprehensively assessed in the ex post evaluation stage. Social media's capability to facilitate interpersonal and group interaction highlights new and unique opportunities for government officials, company managers and researchers to

Acknowledgments

This research work was supported by the National Natural Science Foundation of China (General Program Nos. 51479100, 11272178 and 51379104), and grants (No. 2015-KY-5) from State Key Laboratory of Hydroscience and Engineering, China.

Hanchen Jiang Ph.D. candidate at Department of Hydraulic Engineering in Tsinghua University, China. His major is management science and engineering and his research interests are related to public opinion and general intelligence on controversial issues, such as large infrastructure projects and energy sectors. He is good at using information technologies, such as web crawler and natural language processing methods, to collect and analyze unstructured data which are related to the interaction

References (104)

  • N. Hartley et al.

    Public participation in environmental impact assessment — implementing the Aarhus convention

    Environ. Impact Assess. Rev.

    (2005)
  • W. He et al.

    Social media competitive analysis and text mining: a case study in the pizza industry

    Int. J. Inf. Manag.

    (2013)
  • D. Hidalgo et al.

    TransMilenio BRT system in Bogota, high performance and positive impact — main results of an ex-post evaluation

    Res. Transp. Econ.

    (2013)
  • G. Higgs et al.

    Using IT approaches to promote public participation in renewable energy planning: prospects and challenges

    Land Use Policy

    (2008)
  • S. Jackson et al.

    Resettlement for China's three gorges dam: socio-economic impact and institutional tensions

    Communist Post-Communist Stud.

    (2000)
  • C. Ji et al.

    New Internet search volume-based weighting method for integrating various environmental impacts

    Environ. Impact Assess. Rev.

    (2016)
  • H. Jiang et al.

    Finding academic concerns of the three gorges project based on a topic modeling approach

    Ecol. Indic.

    (2016)
  • A. Jobert et al.

    Local acceptance of wind energy: factors of success identified in French and German case studies

    Energ Policy

    (2007)
  • C.R. Jones et al.

    Understanding ‘local’ opposition to wind development in the UK: how big is a backyard?

    Energ Policy

    (2010)
  • A.P. Kirilenko et al.

    Public microblogging on climate change: one year of twitter worldwide

    Glob. Environ. Chang.

    (2014)
  • A.P. Kirilenko et al.

    People as sensors: mass media and local temperature influence climate change discussion on Twitter

    Glob. Environ. Chang.

    (2015)
  • N. Lapidot-Lefler et al.

    Effects of anonymity, invisibility, and lack of eye-contact on toxic online disinhibition

    Comput. Hum. Behav.

    (2012)
  • G. Lee et al.

    An open government maturity model for social media-based public engagement

    Gov. Inf. Q.

    (2012)
  • T. Li et al.

    Strategic environmental assessment performance factors and their interaction: an empirical study in China

    Environ. Impact Assess. Rev.

    (2016)
  • Z.Z. Liu et al.

    Handling social risks in government-driven mega project: an empirical case study from West China

    Int. J. Proj. Manag.

    (2016)
  • F.D. Musall et al.

    Local acceptance of renewable energy — a case study from Southeast Germany

    Energ Policy

    (2011)
  • H. Nakamura

    The economic evaluation of transport infrastructure: needs for international comparisons

    Transp. Policy

    (2000)
  • S.T. Ng et al.

    A public private people partnerships (P4) process framework for infrastructure development in Hong Kong

    Cities

    (2013)
  • T.H. Nguyen et al.

    Sentiment analysis on social media for stock movement prediction

    Expert Syst. Appl.

    (2015)
  • C. O'Faircheallaigh

    Public participation and environmental impact assessment: purposes, implications, and lessons for public policy making

    Environ. Impact Assess. Rev.

    (2010)
  • N.O. Olsson et al.

    Influence of reference points in ex post evaluations of rail infrastructure projects

    Transp. Policy

    (2010)
  • P.A. Pilavachi et al.

    Ex-post evaluation of European energy models

    Energ Policy

    (2008)
  • J. Schindler et al.

    Sustainability impact assessment to improve food security of smallholders in Tanzania

    Environ. Impact Assess. Rev.

    (2016)
  • B.S. Tang et al.

    Social impact assessment and public participation in China: a case study of land requisition in Guangzhou

    Environ. Impact Assess. Rev.

    (2008)
  • W. Tang et al.

    Risk management of hydropower development in China

    Energy

    (2013)
  • D. Tullos

    Assessing the influence of environmental impact assessments on science and policy: an analysis of the Three Gorges Project

    J. Environ. Manag.

    (2009)
  • O.O. Ugwu et al.

    Sustainability appraisal in infrastructure projects (SUSAIP): Part 1. Development of indicators and computational methods

    Autom. Constr.

    (2006)
  • J.C.J.M. van den Bergh et al.

    Social learning by doing in sustainable transport innovations: ex-post analysis of common factors behind successes and failures

    Res. Policy

    (2007)
  • R. Wüstenhagen et al.

    Social acceptance of renewable energy innovation: an introduction to the concept

    Energ Policy

    (2007)
  • A. Akintoye et al.

    Achieving best value in private finance initiative project procurement

    Constr. Manag. Econ.

    (2003)
  • P. André et al.

    Public participation international best practice principles

  • C. Ashley et al.

    Creative strategies in social media marketing: an exploratory study of branded social content and consumer engagement

    Psychol. Mark.

    (2015)
  • D.R. Becker et al.

    A participatory approach to social impact assessment: the interactive community forum

    Environ. Impact Assess. Rev.

    (2003)
  • D.M. Blei

    Probabilistic topic models

    Commun. ACM

    (2012)
  • D.M. Blei et al.

    A correlated topic model of science

    Ann. Appl. Stat.

    (2007)
  • D.M. Blei et al.

    Latent Dirichlet allocation

    J. Mach. Learn. Res.

    (2003)
  • G. Brown et al.

    Measuring change in place values for environmental and natural resource planning using public participation GIS (PPGIS): results and challenges for longitudinal research

    Soc. Nat. Resour.

    (2014)
  • J.A. Brzozowski

    Derivatives of regular expressions

    J. ACM

    (1964)
  • C. Cao et al.

    Progress in the development of national knowledge infrastructure

    J. Comput. Sci. Technol.

    (2002)
  • J. Cao et al.

    Web-based traffic sentiment analysis: methods and applications

    IEEE Trans. Intell. Transp. Syst.

    (2014)
  • Cited by (67)

    View all citing articles on Scopus

    Hanchen Jiang Ph.D. candidate at Department of Hydraulic Engineering in Tsinghua University, China. His major is management science and engineering and his research interests are related to public opinion and general intelligence on controversial issues, such as large infrastructure projects and energy sectors. He is good at using information technologies, such as web crawler and natural language processing methods, to collect and analyze unstructured data which are related to the interaction among society, environment and engineering projects.

    Maoshan Qiang Professor of Department of Hydraulic Engineering and Department of Construction Management at Tsinghua University, China. His principle research interests are project delivery and organizational structure of construction projects. His current studies focus on project governance, organizational project management, and public opinion analysis, immigration resettlement and social impact assessment of large hydropower projects. He is the author or co-author of > 100 publications. He has over 30-year experience in project management and ex-post assessment consulting for large enterprises and governments.

    Pen Lin Professor of Department of Hydraulic Engineering at Tsinghua University, China. His research interests lie in rock mechanism, construction material, information technologies in construction management and information management of large construction projects. He has published over 100 publications related to hydropower projects. He has a lot of practical experiences in the field of construction management and information technologies in large hydropower projects.

    View full text