Dataset Information

Predicting Depression From Language-Based Emotion Dynamics: Longitudinal Analysis of Facebook and Twitter Status Updates.

ABSTRACT: BACKGROUND:Frequent expression of negative emotion words on social media has been linked to depression. However, metrics have relied on average values, not dynamic measures of emotional volatility. OBJECTIVE:The aim of this study was to report on the associations between depression severity and the variability (time-unstructured) and instability (time-structured) in emotion word expression on Facebook and Twitter across status updates. METHODS:Status updates and depression severity ratings of 29 Facebook users and 49 Twitter users were collected through the app MoodPrism. The average proportion of positive and negative emotion words used, within-person variability, and instability were computed. RESULTS:Negative emotion word instability was a significant predictor of greater depression severity on Facebook (rs(29)=.44, P=.02, 95% CI 0.09-0.69), even after controlling for the average proportion of negative emotion words used (partial rs(26)=.51, P=.006) and within-person variability (partial rs(26)=.49, P=.009). A different pattern emerged on Twitter where greater negative emotion word variability indicated lower depression severity (rs(49)=-.34, P=.01, 95% CI -0.58 to 0.09). Differences between Facebook and Twitter users in their emotion word patterns and psychological characteristics were also explored. CONCLUSIONS:The findings suggest that negative emotion word instability may be a simple yet sensitive measure of time-structured variability, useful when screening for depression through social media, though its usefulness may depend on the social media platform.

SUBMITTER: Seabrook EM

PROVIDER: S-EPMC5964306 | biostudies-literature | 2018 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Depression From Language-Based Emotion Dynamics: Longitudinal Analysis of Facebook and Twitter Status Updates.

Seabrook Elizabeth M EM Kern Margaret L ML Fulcher Ben D BD Rickard Nikki S NS

Journal of medical Internet research 20180508 5

<h4>Background</h4>Frequent expression of negative emotion words on social media has been linked to depression. However, metrics have relied on average values, not dynamic measures of emotional volatility.<h4>Objective</h4>The aim of this study was to report on the associations between depression severity and the variability (time-unstructured) and instability (time-structured) in emotion word expression on Facebook and Twitter across status updates.<h4>Methods</h4>Status updates and depression ...[more]

PMID: 29739736

Similar Datasets

Project description:When the Zika virus outbreak became a global health emergency in early 2016, the scientific community responded with an increased output of Zika-related research. This upsurge in research naturally made its way into academic journals along with editorials, news, and reports. However, it is not yet known how or whether these scholarly communications were distributed to the populations most affected by Zika.To understand how scientific outputs about Zika reached global and local audiences, we collected Tweets and Facebook posts that linked to Zika-related research in the first six months of 2016. Using a language detection algorithm, we found that up to 90% of Twitter and 76% of Facebook posts are in English. However, when none of the authors of the scholarly article are from English-speaking countries, posts on both social media are less likely to be in English. The effect is most pronounced on Facebook, where the likelihood of posting in English is between 11 and 16% lower when none of the authors are from English-speaking countries, as compared to when some or all are. Similarly, posts about papers written with a Brazilian author are 13% more likely to be in Portuguese on Facebook than when made on Twitter.Our main conclusion is that scholarly communication on Twitter and Facebook of Zika-related research is dominated by English, despite Brazil being the epicenter of the Zika epidemic. This result suggests that scholarly findings about the Zika virus are unlikely to be distributed directly to relevant populations through these popular online mediums. Nevertheless, there are differences between platforms. Compared to Twitter, scholarly communication on Facebook is more likely to be in the language of an author's country. The Zika outbreak provides a useful case-study for understanding how scientific outputs are communicated to relevant populations. Our results suggest that Facebook is a more effective channel than Twitter, if communication is desired to be in the native language of the affected country. Further research should explore how local media-such as governmental websites, newspapers and magazines, as well as television and radio-disseminate scholarly publication.

Project description:Health organizations are increasingly using social media, such as Twitter, to disseminate health messages to target audiences. Determining the extent to which the target audience (e.g., age groups) was reached is critical to evaluating the impact of social media education campaigns. The main objective of this study was to examine the separate and joint predictive validity of linguistic and metadata features in predicting the age of Twitter users. We created a labeled dataset of Twitter users across different age groups (youth, young adults, adults) by collecting publicly available birthday announcement tweets using the Twitter Search application programming interface. We manually reviewed results and, for each age-labeled handle, collected the 200 most recent publicly available tweets and user handles' metadata. The labeled data were split into training and test datasets. We created separate models to examine the predictive validity of language features only, metadata features only, language and metadata features, and words/phrases from another age-validated dataset. We estimated accuracy, precision, recall, and F1 metrics for each model. An L1-regularized logistic regression model was conducted for each age group, and predicted probabilities between the training and test sets were compared for each age group. Cohen's d effect sizes were calculated to examine the relative importance of significant features. Models containing both Tweet language features and metadata features performed the best (74% precision, 74% recall, 74% F1) while the model containing only Twitter metadata features were least accurate (58% precision, 60% recall, and 57% F1 score). Top predictive features included use of terms such as "school" for youth and "college" for young adults. Overall, it was more challenging to predict older adults accurately. These results suggest that examining linguistic and Twitter metadata features to predict youth and young adult Twitter users may be helpful for informing public health surveillance and evaluation research.

Project description:BackgroundThe COVID-19 pandemic has necessitated a rapid shift in how individuals interact with and receive fundamental services, including health care. Although telemedicine is not a novel technology, previous studies have offered mixed opinions surrounding its utilization. However, there exists a dearth of research on how these opinions have evolved over the course of the current pandemic.ObjectiveThis study aims to evaluate how the language and sentiment surrounding telemedicine has evolved throughout the COVID-19 pandemic.MethodsTweets published between January 1, 2020, and April 24, 2021, containing at least one telemedicine-related and one COVID-19-related search term ("telemedicine-COVID") were collected from the Twitter full archive search (N=351,718). A comparator sample containing only COVID-19 terms ("general-COVID") was collected and sampled based on the daily distribution of telemedicine-COVID tweets. In addition to analyses of retweets and favorites, sentiment analysis was performed on both data sets in aggregate and within a subset of tweets receiving the top 100 most and least retweets.ResultsTelemedicine gained prominence during the early stages of the pandemic (ie, March through May 2020) before leveling off and reaching a steady state from June 2020 onward. Telemedicine-COVID tweets had a 21% lower average number of retweets than general-COVID tweets (incidence rate ratio 0.79, 95% CI 0.63-0.99; P=.04), but there was no difference in favorites. A majority of telemedicine-COVID tweets (180,295/351,718, 51.3%) were characterized as "positive," compared to only 38.5% (135,434/351,401) of general-COVID tweets (P<.001). This trend was also true on a monthly level from March 2020 through April 2021. The most retweeted posts in both telemedicine-COVID and general-COVID data sets were authored by journalists and politicians. Whereas the majority of the most retweeted posts within the telemedicine-COVID data set were positive (55/101, 54.5%), a plurality of the most retweeted posts within the general-COVID data set were negative (44/89, 49.4%; P=.01).ConclusionsDuring the COVID-19 pandemic, opinions surrounding telemedicine evolved to become more positive, especially when compared to the larger pool of COVID-19-related tweets. Decision makers should capitalize on these shifting public opinions to invest in telemedicine infrastructure and ensure its accessibility and success in a postpandemic world.

Project description:We examined whether self-generated (status updates) or other-generated (wall-posts) information on Facebook influenced the impression formed of the target individual. Along with examining reliance on particular types of information, we explored the valence (positive/ neutral/ negative) of the information, as reliance on self-generated or other-generated information may depend on whether self-presentation is perceived (i.e., presenting oneself positively / not negatively). Self-presentation may be perceived if the targets have positive/ neutral statuses, while negative statuses would indicate a lack of self-presentation. In line with previous research, participants should rely on other-generated information (wall-posts) to form an impression when participants are viewed to have self-presented (positive / neutral status updates), as this information could be viewed as unreliable. Forty participants rated nine Facebook profiles where statuses and wall-posts portrayed personality traits varying in valence. Each profile consisted of a neutral profile photo, three status updates (all positive, negative, or neutral) and three wall-posts (all positive, negative, or neutral). Materials were established in two pilots. Impression formation was measured as perceived social, task, and physical attractiveness of the target individual. Participants also ranked the profiles for likeability. Supporting our expectations, other-generated information (wall-posts) dominated impression formation for social attractiveness when self-generated information (status updates) was positive/ neutral. Task attractiveness was affected by information valence, regardless of source (self or other). Despite the inclusion of neutral photos, physical attractiveness was affected by self-generated information, with negative statuses lowering physical attractiveness. We suggest that these findings have implications for impression formation beyond the Facebook setting. The 557 traits analyzed in Pilot 1 are available as supporting information (S1 Dataset) and may be useful for other impression formation researchers.

Dataset Information

Predicting Depression From Language-Based Emotion Dynamics: Longitudinal Analysis of Facebook and Twitter Status Updates.

Publications

Predicting Depression From Language-Based Emotion Dynamics: Longitudinal Analysis of Facebook and Twitter Status Updates.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets