Dataset Information

Weather impacts expressed sentiment.

ABSTRACT: We conduct the largest ever investigation into the relationship between meteorological conditions and the sentiment of human expressions. To do this, we employ over three and a half billion social media posts from tens of millions of individuals from both Facebook and Twitter between 2009 and 2016. We find that cold temperatures, hot temperatures, precipitation, narrower daily temperature ranges, humidity, and cloud cover are all associated with worsened expressions of sentiment, even when excluding weather-related posts. We compare the magnitude of our estimates with the effect sizes associated with notable historical events occurring within our data.

SUBMITTER: Baylis P

PROVIDER: S-EPMC5918636 | biostudies-literature | 2018

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Weather impacts expressed sentiment.

Baylis Patrick P Obradovich Nick N Kryvasheyeu Yury Y Chen Haohui H Coviello Lorenzo L Moro Esteban E Cebrian Manuel M Fowler James H JH

PloS one 20180425 4

We conduct the largest ever investigation into the relationship between meteorological conditions and the sentiment of human expressions. To do this, we employ over three and a half billion social media posts from tens of millions of individuals from both Facebook and Twitter between 2009 and 2016. We find that cold temperatures, hot temperatures, precipitation, narrower daily temperature ranges, humidity, and cloud cover are all associated with worsened expressions of sentiment, even when exclu ...[more]

PMID: 29694424

Similar Datasets

Project description:BackgroundThe COVID-19 pandemic has had enormous impacts on people's lives, including disruptions to their normal ways of behaving, working, and interacting with others. Understanding and documenting these experiences is important to inform the ongoing response to COVID-19 and disaster preparedness efforts.ObjectiveThe aim of this study was to examine the psychosocial impacts of COVID-19 on a sample of Australian adults.MethodsThe data analyzed were derived from a larger cross-sectional survey of Australian adults that was administered during the month of May 2020. Participants (N=3483) were asked in which ways COVID-19 had most greatly impacted them; the responses produced a text data set containing 1 COVID-19 impact story for each participant, totaling 86,642 words. Participants also completed assessments of their sociodemographic characteristics (sex, age, financial stress), level of concern related to COVID-19, personality trait profile, and satisfaction with life. Impact stories were analyzed using sentiment analysis and compared against the Theoretical Domains Framework to determine the most frequently impacted life domains. Finally, a multinomial regression analysis, stratified by participant sex, was conducted to identify the associations of psychological and demographic socializations with sentiment toward COVID-19.ResultsIn total, 3483 participants completed the survey, the majority of whom were female (n=2793, 80.2%). Participants' impact stories were most commonly categorized as neutral (1544/3483, 44.3%), followed by negative (1136/3483, 32.6%) and positive (802/3483, 23.1%). The most frequently impacted life domains included behavioral regulation, environmental context and resources, social influences, and emotions, suggesting that the COVID-19 pandemic was impacting these areas of participants' lives the most. Finally, the regression results suggested that for women, lower satisfaction with life and higher financial stress were associated with increased likelihood of negative, rather than positive, sentiment (P<.001); however, the proportion of variance in the sentiment that was explained was very small (<5%).ConclusionsParticipant sentiment toward COVID-19 varied. High rates of neutral and negative sentiment were identified. Positive sentiment was identified but was not as common. Impacts to different areas of people's lives were identified, with a major emphasis on behavioral regulation and related domains such as social influences, environmental context and resources, and emotions. Findings may inform the development of mental health and social support resources and interventions to help alleviate the psychosocial consequences of disaster response measures.

Dataset Information

Weather impacts expressed sentiment.

Publications

Weather impacts expressed sentiment.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets