Nothing Special   »   [go: up one dir, main page]

Published November 1, 2018 | Version v1
Dataset Open

The relationship between linguistic expression and symptoms of depression, anxiety, and suicidal thoughts: A longitudinal study of blog content

  • 1. Black Dog Institute, UNSW Sydney, Sydney, Australia
  • 2. Centre for Pattern Recognition and Data Analytics, Deakin University, Australia

Description

To investigate the associations between linguistic features and symptoms of depression, generalised anxiety, and suicidal ideation, we extracted linguistic features from individuals’ blog content and correlated it with validated mental health data in a longitudinal study (n=38). Depressive symptoms were assessed using the self-report Patient Health Questionnaire (PHQ-9), anxiety symptoms using the self-report Generalised Anxiety Disorder Scale (GAD-7), and social media data was analysed using the Linguistic Inquiry and Word Count (LIWC) tool for linguistic features. Bivariate and multivariate analyses were performed to investigate the correlations between the linguistic features and mental health scores between subjects. We then used the multivariate regression model to predict longitudinal changes in mood within subjects.

Notes

HC and this research were financially supported by NHMRC John Cade Fellowship APP1056964. BOD and MEL were supported by the Society for Mental Health Research (SMHR) Early Career Researcher Awards. TB was supported by a NARSAD Young Investigator Grant from the Brain & Behavior Research Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Files

groundtruth_individualdata.csv

Files (185.9 kB)

Name Size Download all
md5:a5e194e8216dade31c8cc97bf7fb2fb1
148.2 kB Preview Download
md5:7fb964875f0ec78c45d464f19707eb72
37.7 kB Preview Download

Additional details

Funding

Prevention of depression using e health technologies 1056964
National Health and Medical Research Council