Elsevier

Technology in Society

Volume 67, November 2021, 101724
Technology in Society

A review of social media-based public opinion analyses: Challenges and recommendations

https://doi.org/10.1016/j.techsoc.2021.101724Get rights and content

Highlights

  • A review on studies on social media-based public opinion analyses is presented.

  • Challenges in data collection, data quality, and data mining are discussed.

  • Promising directions of the research stream are presented.

  • Ethical considerations in social media-based public opinion analysis are highlighted.

  • This study contributes to public opinion analysis and social media use.

Abstract

Compared with survey polls, social media can yield a better and more comprehensive understanding of public perceptions of special topics in a more scientific manner. However, despite this advantage, there seem to be limited investigations into the challenges in social media-based public opinion analysis. This study offers an understanding of the challenges in this field and some corresponding recommendations. Through a systematic literature review, we identify 54 papers to analyze and discuss issues related to data collection, data quality, and data mining. This paper summarizes a framework for social media-based public opinion analysis as well as the commonly employed data mining methodologies. We found that collecting public opinion data from Facebook and Weibo is difficult because of their restricted application programming interface and measures against Web Crawler. How to effectively and conveniently delete invalid data and how to design data mining methods for social media data, especially for those in Chinese, are still two main challenges in social media-based public opinion analysis. We claim that using multiple data sources, optimizing keyword settings, enhancing interdisciplinary cooperation, and paying more attention to the functional role of social media can benefit the development of social media-based public opinion analysis. This study also highlights the potential risks of releasing the personal information of the public in the use of social media data in research.

Introduction

Public opinions about policies and events have attracted the attention of scholars and policymakers for a long time [1,2], and they are considered important evidence in making and adjusting decisions [3,4] and an essential concept in the underpinnings of democracy [5]. Lippmann, one of the founders of communication science, published a book entitled Public Opinion, in which public opinion was defined as an organic product formed in the process of community discussion [6]. He expounded on the internal and external factors in the formation of public opinions. His book can be considered the beginning of public opinion analysis. In general, existing investigations into public opinion generally began with data collection based on survey polls and then analyzed the data using traditional qualitative and quantitative methods [7,8]. A standard survey poll is generated through two processes of selecting representative samples and surveying the samples about their opinions [5]. However, some scholars argue that a survey poll is more likely a way of representing a small group of individuals’ viewpoints rather than public opinion or social opinion [9,10]. In addition, the effects of the hierarchical nature of society on the role of elites in the formation of opinions are considerable in the survey process [11]. Finally, a general research paradigm of survey polls has been built in which the opinions of a few subjectively selected individuals are used to measure the public as a whole. This paradigm largely restricts the publicness of public opinion by representing it based on private and subjective settings. Moreover, due to the time required for and high capital costs involved in data collection for survey polls, data quantity is usually limited, restricting the representativeness of the findings [12] to a large extent. In recent years, with the popularity of cell phones and internet use, cell phone and web polling have emerged. These new forms of opinion collection approaches can enlarge the size of the data set collected, expand sample diversity, reduce associated costs, and accelerate the speed of data collection [13]. Nevertheless, the disadvantages of representative sampling and subjectivity in sample selection remain. Reliability and accuracy problems are also frequently questioned [5].

Social media platforms provide a new way of representing and measuring public opinions. There has been a significant increase in the adoption and use of social media by both the general public and particular subpopulations, such as government sectors, enterprises, and celebrities [14]. In past decades, social media-based public opinion (SMPO) analysis has been conducted in various fields, including social science [15], politics [16], education [17], medical science [18,19], marketing [20], transportation [21], finance [22], knowledge sharing [23], and disaster management [24,25], showing high interest and considerable effectiveness. A specific form in which users both access and share opinions and perceptions is showing a greater transition in social media platforms than perhaps at any previous point in history. A huge amount of opinion data is freely available to researchers, and it shows a more comprehensive picture with higher cost efficiency. Using social media as a data source for public opinion collection can subvert some of the underlying methodological limitations of traditional surveys [10,26]. For example, representative sampling [27,28], the hierarchical nature of opinion formation [5], and difficulties in obtaining time-series data [29] can be eliminated by using social media. Anstead and O'Loughlin [16] suggested that a public and collective stage has been formed in social media mainly because of its conversational nature. McGregor [5] found that social media offers a more temporal-sensitive channel for obtaining and gauging public sentiment toward particular policies and events as the opinions posted when a policy is released and an event occurs are recorded on platforms and cannot be changed. Salleh [30] suggested that social media can not only show a better understanding of public perception in a more scientific manner but can also be helpful in forecasting future political trends and shaping society's worldview. These advantages motivate scholars to use social media as a data source instead of survey polls.

However, despite the advantages brought about by the use of social media in collecting public opinions, certain problems should be taken into consideration, such as difficulties in guaranteeing the relativity of the data, barriers to sharing information honestly and openly, and the reliability and validity of data reprocessing. Although previous studies have more or less proposed some valuable findings, there is still a lack of systematic research on the challenges in the use of social media for public opinion analysis. Therefore, this paper aims to identify the main challenges in SMPO analysis by reviewing existing studies and proposing some practical recommendations.

The research questions are as follows:

  • (i)

    What is the current state of research on SMPO analysis?

  • (ii)

    What are the key challenges identified in previous studies?

  • (iii)

    What are the future perspectives of SMPO analysis?

We answer these questions through a literature review in which we investigate how social media data are used in public opinion analysis with the aim of proposing a framework and some practical suggestions that can guide and help scholars conduct SMPO analysis in a more standardized and comprehensive manner. For this purpose, the study presents a systematic review of the current studies on SMPO analysis by analyzing 54 papers. We identify the motivation for the adoption of social media in public opinion analysis and summarize the existing primary challenges from the perspectives of data collection, data quality, data mining, and ethical considerations. This review also suggests some potential areas where future research can improve upon the significance of the study.

Section snippets

Methodology

To gain a comprehensive understanding of the way in which social media contributes to public opinion analysis, a systematic review of the related literature was carried out in this study, enabling us to identify the existing challenges in this application. We used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol to conduct this review [31]. A qualitative meta-synthesis of the set of publications was conducted to answer the research questions [32], which

Characteristics of the studies

This section presents the following characteristics of the studies reviewed: date of publication, keywords, data source, and design framework for SMPO analysis.

Challenges in and recommendations on social media adoption in public opinion analysis

Based on the reviewed studies, several challenges in SMPO analysis are proposed in this section by considering data collection, data quality, data mining, and ethical considerations.

Conclusion

SMPO analysis is expected to become a central branch of research in future studies as it provides a new way for scholars to better and more comprehensively understand public perceptions on particular issues. This study conducted a systematic review of the corpus of literature on SMPO analysis. On the basis of the reviewed studies, we summarized a general framework for SMPO analysis and identified some challenges in this field by considering data collection, data quality, data mining, and

Declaration of conflicting interests

The authors declare no conflict of interest.

Ethic consideration

This study was performed in accordance with the ethical guidelines from the Ethics Committee of Beijing University of Technology. All subjects gave their written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Ethics Committee of Beijing University of Technology.

Acknowledgement

This research was supported by the National Natural Science Foundation of China [grant number 71904010], the Fundamental Research Funds for the Central Universities [CUC210C002], and the Funding Project of Interdisciplinary Research Institute of Beijing University of Technology [grant number 2021110115].

References (160)

  • W. Liu et al.

    Developing a multi-level organization-public dialogic communication framework to assess social media-mediated disaster communication and engagement outcomes

    Publ. Relat. Rev.

    (2020)
  • N. Pourebrahim

    Understanding communication dynamics on Twitter during natural disasters: a case study of Hurricane Sandy

    International Journal of Disaster Risk Reduction

    (2019)
  • T. Vepsäläinen et al.

    Facebook likes and public opinion: predicting the 2015 Finnish parliamentary elections

    Govern. Inf. Q.

    (2017)
  • N.F. Ibrahim et al.

    Decoding the sentiment dynamics of online retailing customers: time series analysis of social media

    Comput. Hum. Behav.

    (2019)
  • R. Dekker et al.

    Social Media Adoption in the Police: Barriers and Strategies

    (2020)
  • C.C. Flores et al.

    Twitter information for contributing to the strategic digital city: towards citizens as co-managers

    Telematics Inf.

    (2018)
  • H. Yu

    Global science discussed in local altmetrics: weibo and its comparison with Twitter

    Journal of Informetrics

    (2017)
  • L. Zheng et al.

    Innovation through social media in the public sector: information and interactions

    Govern. Inf. Q.

    (2014)
  • Y. Lyu et al.

    Exploring public attitudes of child abuse in mainland China: a sentiment analysis of China's social media Weibo

    Child. Youth Serv. Rev.

    (2020)
  • P. Shao et al.

    How does social media change Chinese political culture? The formation of fragmentized public sphere

    Telematics Inf.

    (2017)
  • X. Lin et al.

    Exploring extreme events on social media: a comparison of user reposting/retweeting behaviors on Twitter and Weibo

    Comput. Hum. Behav.

    (2016)
  • J.W. Ma et al.

    A window to the ideal self: a study of UK Twitter and Chinese Sina Weibo selfie-takers and the implications for marketers

    J. Bus. Res.

    (2017)
  • B. Li

    Why we follow: examining motivational differences in following sport organizations on Twitter and Weibo

    Sport Manag. Rev.

    (2019)
  • M.J. Pelletier et al.

    Fexit: the effect of political and promotional communication from friends and family on Facebook exiting intentions

    J. Bus. Res.

    (2021)
  • C. Kudeshia et al.

    Spreading love through fan page liking: a perspective on small scale entrepreneurs

    Comput. Hum. Behav.

    (2016)
  • Y. Yan et al.

    Mining public sentiments and perspectives from geotagged social media data for appraising the post-earthquake recovery of tourism destinations

    Appl. Geogr.

    (2020)
  • B. Qi et al.

    A Framework with Efficient Extraction and Analysis of Twitter Data for Evaluating Public Opinions on Transportation Services

    (2020)
  • P. Ji

    Emotional criticism as public engagement: how weibo users discuss "Peking University statues wear face-masks

    Telematics Inf.

    (2016)
  • J.D. Featherstone

    Exploring childhood vaccination themes and public opinions on Twitter: a semantic network analysis

    Telematics Inf.

    (2020)
  • F. Didegah et al.

    Investigating the quality of interactions and public engagement around scientific papers on Twitter

    Journal of Informetrics

    (2018)
  • M. Cho et al.

    Public engagement with nonprofit organizations on Facebook

    Publ. Relat. Rev.

    (2014)
  • K. Mora

    Public perceptions of building seismic safety following the Canterbury earthquakes: a qualitative analysis using Twitter and focus groups

    International Journal of Disaster Risk Reduction

    (2015)
  • J. Benthaus et al.

    Social media management strategies for organizational impression management and their effect on public perception

    J. Strat. Inf. Syst.

    (2016)
  • X. Cao

    Using Twitter to better understand the spatiotemporal patterns of public sentiment: a case study in Massachusetts, USA

    Int. J. Environ. Res. Publ. Health

    (2018)
  • B. Cai

    How scholars and the public perceive a "low carbon city" in China

    J. Clean. Prod.

    (2017)
  • X. Liu et al.

    Attention and sentiment of Chinese public toward green buildings based on Sina Weibo

    Sustainable Cities and Society

    (2019)
  • L. Su et al.

    Online public response to a service failure incident: implications for crisis communications

    Tourism Manag.

    (2019)
  • E. D'Andrea

    Monitoring the public opinion about the vaccination topic from tweets analysis

    Expert Syst. Appl.

    (2019)
  • S. Poria

    Ensemble application of convolutional neural networks and multiple kernel learning for multimodal sentiment analysis

    Neurocomputing

    (2017)
  • H.E. Krugman

    The impact of television advertising: learning without involvement

    Publ. Opin. Q.

    (1965)
  • Alan

    An analysis of the relationship between news coverage of health topics and public opinion of the most important health problems in the United States

    J. Health Educ.

    (1992)
  • E. Noelle-Neumann

    Public opinion and the classical tradition: a re-evaluation

    Publ. Opin. Q.

    (1979)
  • S.C. McGregor

    Social media as public opinion: how journalists use social media to represent public opinion

    Journalism

    (2019)
  • W. Lippmann

    Public Opinion

    (1922)
  • C. Tobin et al.

    A review of public opinion towards alcohol controls in Australia

    BMC Publ. Health

    (2011)
  • J.S. Fishkin

    Beyond polling alone: the quest for an informed public

    Crit. Rev.

    (2006)
  • Murphy

    Social media in public opinion research executive summary of the AAPOR task force on emerging technologies in public opinion research[J]

    Publ. Opin. Q.

    (2014)
  • B. Jiang

    Mining twitter to assess the public perception of the "internet of things

    PloS One

    (2016)
  • M.H. Huang et al.

    The internet, social capital, and civic engagement in Asia

    Soc. Indicat. Res.

    (2017)
  • A. Nick et al.

    Social media analysis and public opinion: the 2010 UK general election

    J. Computer-Mediated Commun.

    (2015)
  • Cited by (46)

    View all citing articles on Scopus
    View full text