Dataset Information

Measuring relative opinion from location-based social media: A case study of the 2016 U.S. presidential election.

ABSTRACT: Social media has become an emerging alternative to opinion polls for public opinion collection, while it is still posing many challenges as a passive data source, such as structurelessness, quantifiability, and representativeness. Social media data with geotags provide new opportunities to unveil the geographic locations of users expressing their opinions. This paper aims to answer two questions: 1) whether quantifiable measurement of public opinion can be obtained from social media and 2) whether it can produce better or complementary measures compared to opinion polls. This research proposes a novel approach to measure the relative opinion of Twitter users towards public issues in order to accommodate more complex opinion structures and take advantage of the geography pertaining to the public issues. To ensure that this new measure is technically feasible, a modeling framework is developed including building a training dataset by adopting a state-of-the-art approach and devising a new deep learning method called Opinion-Oriented Word Embedding. With a case study of tweets on the 2016 U.S. presidential election, we demonstrate the predictive superiority of our relative opinion approach and we show how it can aid visual analytics and support opinion predictions. Although the relative opinion measure is proved to be more robust than polling, our study also suggests that the former can advantageously complement the latter in opinion prediction.

SUBMITTER: Gong Z

PROVIDER: S-EPMC7244148 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Measuring relative opinion from location-based social media: A case study of the 2016 U.S. presidential election.

Gong Zhaoya Z Cai Tengteng T Thill Jean-Claude JC Hale Scott S Graham Mark M

PloS one 20200522 5

Social media has become an emerging alternative to opinion polls for public opinion collection, while it is still posing many challenges as a passive data source, such as structurelessness, quantifiability, and representativeness. Social media data with geotags provide new opportunities to unveil the geographic locations of users expressing their opinions. This paper aims to answer two questions: 1) whether quantifiable measurement of public opinion can be obtained from social media and 2) wheth ...[more]

PMID: 32442212

Similar Datasets

Project description:ImportanceIn the U.S. presidential election of 2016, substantial shift in voting patterns occurred relative to previous elections. Although this shift has been associated with both education and race, the extent to which this shift was related to public health status is unclear.ObjectiveTo determine the extent to which county community health was associated with changes in voting between the presidential elections of 2016 and 2012.DesignEcological study with principal component analysis (PCA) using principal axis method to extract the components, then generalized linear regression.SettingGeneral community.ParticipantsAll counties in the United States.ExposuresPhysically unhealthy days, mentally unhealthy days, percent food insecure, teen birth rate, primary care physician visit rate, age-adjusted mortality rate, violent crime rate, average health care costs, percent diabetic, and percent overweight or obese.Main outcomeThe percentage of Donald Trump votes in 2016 minus percentage of Mitt Romney votes in 2012 ("net voting shift").ResultsComplete public health data was available for 3,009 counties which were included in the analysis. The mean net voting shift was 5.4% (+/- 5.8%). Of these 3,009 counties, 2,641 (87.8%) had positive net voting shift (shifted towards Trump) and 368 counties (12.2%) had negative net voting shift (shifted away from Trump). The first principal component ("unhealthy score") accounted for 68% of the total variance in the data. The unhealthy score included all health variables except primary care physician rate, violent crime rate, and health care costs. The mean unhealthy score for counties was 0.39 (SD 0.16). Higher normalized unhealthy score was associated with positive net voting shift (22.1% shift per unit unhealthy, p < 0.0001). This association was stronger in states that switched Electoral College votes from 2012 to 2016 than in other states (5.9% per unit unhealthy, p <0.0001).Conclusions and relevanceSubstantial association exists between a shift toward voting for Donald Trump in 2016 relative to Mitt Romney in 2012 and measures of poor public health. Although these results do not demonstrate causality, these results suggest a possible role for health status in political choices.

Dataset Information

Measuring relative opinion from location-based social media: A case study of the 2016 U.S. presidential election.

Publications

Measuring relative opinion from location-based social media: A case study of the 2016 U.S. presidential election.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets