R Code for Analyzing Youth Crisis Texts After Hurricanes in Louisiana

R code used to analyze 2,149 anonymized crisis text conversations from Louisiana youth between 2018 and 2023. The analysis applied structural topic modeling to examine thematic patterns and temporal trajectories following six hurricanes, including Hurricane Ida. The work was authored by Margaret M. Sugg and last updated on May 12, 2026.

Use Cases

• Replicate structural topic modeling analysis based on the described methodology for crisis text data
• Study temporal patterns of mental health help-seeking based on the defined recovery periods (days -1 to >365)
• Analyze compound disaster effects (hurricane + COVID-19) on crisis topics like Grief and Abuse
• Investigate the 'three waves' pattern of post-disaster psychological response described in the summary

Strengths

• Analysis is based on 2,149 real-world crisis text conversations
• Data covers a multi-year period from 2018 to 2023
• Methodology details are provided, including the use of a temporal proximity algorithm and classification into eight recovery periods
• Specific findings are reported, such as Suicide Ideation increasing from 5.1% to 8.9% at 61-90 days post-landfall

Limitations

• The dataset is the R code (10.5 KB), not the underlying text data; the actual conversation data is not included
• Row count for the source data is unknown, which may limit suitability assessment
• Column-level documentation for any potential accompanying data is absent; field semantics must be inferred from the code

Provenance

Source

figshare

Collection Method

Analysis of anonymized crisis text conversations from Louisiana youth.

Time Range

2018-2023

Freshness

Last updated 2026-05-12 17:41:46

Geography

Louisiana, USA