Network analysis of multivariate data in psychological science

Borsboom, Denny; Deserno, Marie K.; Rhemtulla, Mijke; Epskamp, Sacha; Fried, Eiko I.; McNally, Richard J.; Robinaugh, Donald J.; Perugini, Marco; Dalege, Jonas; Costantini, Giulio; Isvoranu, Adela‐Maria; Wysocki, Anna; Borkulo, Claudia D. van; Bork, Riet van; Waldorp, Lourens J.

doi:10.1038/s43586-021-00055-w

Cited by 421 publications

(352 citation statements)

References 141 publications

Supporting

Mentioning

328

Contrasting

Unclassified

Order By: Relevance

“…To better gauge the relationship between goal adjustment, self-control, and boredom beyond bivariate correlations of aggregate scores, we turned to psychometric network analysis (Borsboom et al, 2021;Christensen et al, 2020). This approach allowed us to estimate a parsimonious network of the constructs in which all individual items are represented by nodes.…”

Section: Psychometric Network Analysismentioning

confidence: 99%

Getting Trapped in a Dead End? Trait Self-Control and Boredom Are Linked to Goal Adjustment

Bieleke¹,

Keller²,

Wolff³

2021

Preprint

View full text Add to dashboard Cite

Disengaging from unattainable goals and reengaging in alternative goals is essential for effective goal pursuit; yet, surprisingly little is known about associated personality factors. Here, we focused on individual differences in self-control (domain-general self-control, if-then planning) and boredom (boredom proneness, boredom avoidance and escape tendencies). Concerning goal adjustment in everyday life (Study 1; N = 323 crowdworkers), if-then planning was associated with worse disengagement and better reengagement. While boredom proneness was associated with poorer reengagement, boredom avoidance and escape tendencies were associated with better reengagement. When goal striving was thwarted during the COVID-19 pandemic (Study 2; N = 97 students), similar associations emerged along with links to anxiety and depression. However, disengagement was no longer associated with if-then planning but instead with better self-control and higher boredom proneness. These results show differential relationships of goal disengagement and reengagement with self-control and boredom, paving the way to a better understanding of who struggles or shines when effective goal adjustment is required.

show abstract

Section: Psychometric Network Analysismentioning

confidence: 99%

Getting Trapped in a Dead End? Trait Self-Control and Boredom Are Linked to Goal Adjustment

Bieleke¹,

Keller²,

Wolff³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In particular, little is known about the relationships between the proposed symptoms. Our study responds to this need from the perspective of network analysis (see, e.g., Borsboom et al, 2021), which is well suited for exploring the structures of relationships between symptomsthose which have been included to and non-included to diagnostic manuals (Fried et al, 2016;Hansen et al, 2021). Previously, some symptoms have been labeled as "core" and other as "peripheral," however, the literature on IGD/GD is mixed regarding which should belong to each class (e.g., Ballou & van Rooij 2021;Snodgrass et al, 2018).…”

Section: Deceptionmentioning

confidence: 88%

Network Structures of Internet Gaming Disorder and Gaming Disorder: Symptom Operationalization Causes Variation

Adamkovič¹,

Martončik²,

Karhulahti³

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: From 2022, the ICD-11 includes the first mental disorder based on digital technology, “gaming disorder”, which was previously suggested as a condition for further study in the DSM-5 (2013). In this study, we provide the first large-scale network analysis of various symptom structures for these constructs to understand the complex interconnections between their proposed symptoms.Methods: Culturally diverse samples of 2,846 digital game players (M = 25.3 years) and 746 esports players (M = 23.5 years) were recruited. Network approach was applied to explore a multiverse of gaming disorder symptom structures, effects of item operationalization, and possible external moderators.Findings: Two symptoms (lack of control and continued use despite problems) present in both, the DSM-5 and ICD-11, were systematically central to most of the analyzed networks. Alternative operationalizations of single items systematically caused significant network differences. Networks were invariant across groups of play style, age, gender, gaming time, and most of the psychosocial characteristics.Interpretation: Our results caution practitioners and researchers when studying and interpreting gaming disorder symptoms. The data indicate that even minor item-level changes can lead to significant network-level changes, thus highlighting the need for careful operationalization. Considering this high sensitivity of the network structure, future clinical validation studies should not be strictly limited to the wordings, criteria, and symptoms of current diagnostic manuals.

show abstract

“…Thus, we selected instruments that assess many different aspects of psychopathology and quality of life in a dimensional/continuous way. While network estimation approaches to handle categorical and ordinal data exists, such data types are still regarded as suboptimal [36]. Moreover, the selection of instruments was driven by the limited time provided by the schools to complete the entire questionnaire battery (max.…”

Section: Instrumentsmentioning

confidence: 99%

Psychopathological Symptoms and Well-Being in Overweight and Underweight Adolescents: A Network Analysis

et al. 2021

View full text Add to dashboard Cite

Overweight and underweight adolescents have an increased risk of psychological problems and reduced quality of life. We used a network analysis approach on a variety of psychopathology and well-being variables to identify central factors in these populations. The network analysis was conducted on data of 344 overweight adolescents (>90th BMI-percentile) and 423 underweight adolescents (<10th BMI-percentile) drawn from a large community sample (10–19 years) including behavioral and emotional problems (Youth Self-Report), eating disorder risk (SCOFF) and well-being variables (KIDSCREEN). Additionally, psychopathology and well-being scores of overweight and underweight individuals were compared with 1.560 normal weight adolescents. Compared to their normal weight peers, overweight adolescents showed elevated psychopathology and eating disorder risk as well as reduced well-being. Underweight adolescents reported increased levels of internalizing problems but no increased eating disorder risk or reduced well-being. The network analysis revealed that anxious/depressed mood and attention problems were the most central and interconnected nodes for both overweight and underweight subsamples. Among underweight individuals, social problems and socially withdrawn behavior additionally functioned as a bridge between other nodes in the network. The results support psychological interventions focusing on improving mood, coping with negative emotions and tackling inner tension.

show abstract

Network analysis of multivariate data in psychological science

Abstract: Characterization of the global network topology and the position of individual nodes in that topology. Psychometric network analysisThe analysis of multivariate psychometric data using network structure estimation and network description.

Cited by 421 publications

References 141 publications

Getting Trapped in a Dead End? Trait Self-Control and Boredom Are Linked to Goal Adjustment

Getting Trapped in a Dead End? Trait Self-Control and Boredom Are Linked to Goal Adjustment

Network Structures of Internet Gaming Disorder and Gaming Disorder: Symptom Operationalization Causes Variation

Psychopathological Symptoms and Well-Being in Overweight and Underweight Adolescents: A Network Analysis

Contact Info

Product

Resources

About