The use of text‐based responses to improve our understanding and prediction of suicide risk

Jacobucci, Ross; Ammerman, Brooke A.; Wilcox, Kenneth Tyler

doi:10.1111/sltb.12668

Cited by 12 publications

(4 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The supervised LDA permits the user to label each document with known properties that can be used for model fitting. Jacobucci et al [ 144 ] provided a recent example of supervised LDA, where they included information on whether the author of each document used in their model had a known history of suicide risk. The study by Šperková [ 145 ] provides further information about variations of LDA (eg, sentiment LDA and factorial LDA).…”

Section: Discussionmentioning

confidence: 99%

Examining Analytic Practices in Latent Dirichlet Allocation Within Psychological Science: Scoping Review

Hagg¹,

Merkouris²,

O’Dea³

et al. 2022

J Med Internet Res

View full text Add to dashboard Cite

Background Topic modeling approaches allow researchers to analyze and represent written texts. One of the commonly used approaches in psychology is latent Dirichlet allocation (LDA), which is used for rapidly synthesizing patterns of text within “big data,” but outputs can be sensitive to decisions made during the analytic pipeline and may not be suitable for certain scenarios such as short texts, and we highlight resources for alternative approaches. This review focuses on the complex analytical practices specific to LDA, which existing practical guides for training LDA models have not addressed. Objective This scoping review used key analytical steps (data selection, data preprocessing, and data analysis) as a framework to understand the methodological approaches being used in psychology research using LDA. Methods A total of 4 psychology and health databases were searched. Studies were included if they used LDA to analyze written words and focused on a psychological construct or issue. The data charting processes were constructed and employed based on common data selection, preprocessing, and data analysis steps. Results A total of 68 studies were included. These studies explored a range of research areas and mostly sourced their data from social media platforms. Although some studies reported on preprocessing and data analysis steps taken, most studies did not provide sufficient detail for reproducibility. Furthermore, the debate surrounding the necessity of certain preprocessing and data analysis steps is revealed. Conclusions Our findings highlight the growing use of LDA in psychological science. However, there is a need to improve analytical reporting standards and identify comprehensive and evidence-based best practice recommendations. To work toward this, we developed an LDA Preferred Reporting Checklist that will allow for consistent documentation of LDA analytic decisions and reproducible research outcomes.

show abstract

Section: Discussionmentioning

confidence: 99%

Examining Analytic Practices in Latent Dirichlet Allocation Within Psychological Science: Scoping Review

Hagg¹,

Merkouris²,

O’Dea³

et al. 2022

J Med Internet Res

View full text Add to dashboard Cite

show abstract

“…These audio check-ins have high potential as a rich source of information about participants' daily experiences. Natural language processing of transcripts of their check-ins can provide a novel window into their thoughts [68][69][70][71]. These audio check-ins provided participants the opportunity to share more openly and candidly (i.e., without close-ended questions) their thoughts, feelings, and progress towards recovery without being limited to researcher-selected prompts.…”

Section: Important Individual Differences In Subjective Perceptions E...mentioning

confidence: 99%

Acceptability of Personal Sensing Among People with Alcohol Use Disorder: Observational Study

Wyant¹,

Moshontz

Ward³

et al. 2022

Preprint

View full text Add to dashboard Cite

Personal sensing may improve digital therapeutics for mental health care by facilitating early screening, symptom monitoring, risk prediction, and personalized/adaptive interventions. However, further development and use of personal sensing first requires better understanding of its acceptability to people targeted for these mental health applications. We assessed the acceptability of both active and passive personal sensing methods in a sample of people with moderate to severe alcohol use disorder using both behavioral and self-report measures. Participants (N = 154; 50% female; mean age = 41; 87% White, 97% Non-Hispanic) in early recovery (1 – 8 weeks of abstinence) from alcohol use disorder were recruited from the Madison, WI area to participate in a 3-month longitudinal study. Participants engaged with active (ecological momentary assessment; EMA, audio check-in, and sleep quality) and passive (geolocation, cellular communication logs, and text message content) personal sensing methods. We assessed 3 behavioral indicators of acceptability: participants’ choices about their participation in the study at various stages in the study procedure, their choice to opt-in to provide data for each personal sensing method, and their compliance for a subset of the active methods (EMA, audio check-in). We also assessed 3 self-report measures of acceptability (interference, dislike, and willingness to use for 1 year) for each method. All but 1 of the individuals who were eligible to participate consented to the personal sensing procedures. Most of these individuals (88%) also returned 1 week later to formally enroll in the study and begin to provide these data. All participants (100%) opted-in to provide data for EMA, sleep quality, and all passive methods (geolocation, cellular communication logs, text message content). Three participants (2%) did not provide any audio check-ins while on study. The average completion rate for all requested EMAs (4X daily) was 81% for 4x daily and 94% for 1x daily. The completion rate for the daily audio check-in was 55%. Aggregate participant ratings indicated all personal sensing methods to be significantly more acceptable (all P’s < .05) compared to neutral across subjective measures of interference, dislike, and willingness to use for 1 year. Participants did not significantly differ in their dislike of active compared to passive methods (P = .23). However, participants reported a higher willingness to use passive methods for 1 year compared to active methods (P = .04). The results of our study suggest that both active and passive personal sensing methods are generally acceptable to people with alcohol use disorder over a longer period than has previously been assessed. This was true even for data streams that contained potentially more sensitive information (e.g., geolocation, cellular communications). Important individual differences were observed both across people and methods which indicate opportunities for future improvements.

show abstract

“…This approach affords the application of text-based analyses, which have the flexibility to highlight the information deemed important by those with lived experience. Such analysis has demonstrated initial promise within the suicide literature (e.g., Jacobucci et al, 2021) and thus has a natural extension to understanding NSSI risk.…”

Section: Introductionmentioning

confidence: 99%

A Text Mining Approach to Characterizing Interpersonal Stress among Individuals with a Nonsuicidal Self-Injury History

Ammerman

Wilcox

Jacobucci

et al. 2023

Preprint

View full text Add to dashboard Cite

Objective: Interpersonal difficulties are salient among those with a history of NSSI and precede NSSI urges and behaviors. Yet, limited research has focused on identifying which aspects of interpersonal stress may confer risk for NSSI. Method: The current study aimed to leverage data from two samples (combined n=206; n=114 with NSSI history) of participant-driven interviews regarding a recent interpersonal stressor to enhance the field’s knowledge of interpersonal difficulties in relation to NSSI risk. Results: Using topic modeling to extract thematic information, analyses identified four main topics: daily difficulties; family members; adjectives/verbal fillers; and friendship/romantic relationships. Relationships between the topics and three predictors (i.e., NSSI history, emotion dysregulation, sample) were examined. In one sample, the proportion of ‘adjectives/verbal fillers’ was greater for participants with a NSSI history and at higher levels of emotion dysregulation. Across samples, for participants with a NSSI history, ‘adjectives/verbal fillers’ and ‘friendship/romantic partners’ increased with levels of emotion dysregulation. Conclusion: Findings highlight a greater use of adjectives and verbal fillers among individuals with a history of NSSI and higher levels of emotion dysregulation. This pattern of language may serve as an indicator of a specific aspect of emotion regulation difficulties that confers risk for NSSI.

show abstract

The use of text‐based responses to improve our understanding and prediction of suicide risk

Cited by 12 publications

References 45 publications

Examining Analytic Practices in Latent Dirichlet Allocation Within Psychological Science: Scoping Review

Examining Analytic Practices in Latent Dirichlet Allocation Within Psychological Science: Scoping Review

Acceptability of Personal Sensing Among People with Alcohol Use Disorder: Observational Study

A Text Mining Approach to Characterizing Interpersonal Stress among Individuals with a Nonsuicidal Self-Injury History

Contact Info

Product

Resources

About