Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Depression, ångest och stress hos universitetsstudenter: en tvärsnittsstudie med nätverks- och LASSO-baserad analys av PHQ-9, GAD-7 och PSS-10

Bernborg, Emma LU and Watts, Elliot LU (2026) PSYK12 20252
Department of Psychology
Abstract
Depression, anxiety, and stress are closely related forms of psychological distress that often exhibit substantial symptomatic overlap. The present study examined how individual items within the self-report instruments PHQ-9, GAD7, and PSS-10 were related to their respective scales as well as to each other
among university students, with a focus on which items appeared most central and indicative of variation in levels of depression, anxiety, and stress. The analyses were based on large, openly available datasets comprising over 50,000 university students from multiple countries. The analyses included descriptive statistics, item–total correlations, estimation of internal consistency (Cronbach’s α), confirmatory factor analysis,... (More)
Depression, anxiety, and stress are closely related forms of psychological distress that often exhibit substantial symptomatic overlap. The present study examined how individual items within the self-report instruments PHQ-9, GAD7, and PSS-10 were related to their respective scales as well as to each other
among university students, with a focus on which items appeared most central and indicative of variation in levels of depression, anxiety, and stress. The analyses were based on large, openly available datasets comprising over 50,000 university students from multiple countries. The analyses included descriptive statistics, item–total correlations, estimation of internal consistency (Cronbach’s α), confirmatory factor analysis, regularized LASSO regressions, and network analyses at the item level. The results showed that all instruments exhibited clear item-specific patterns, with certain symptoms showing stronger associations
with other items than others. Network analyses indicated that affective and cognitive symptoms often occupied more central positions in the networks, whereas somatic symptoms were generally more peripheral. Overall, the study demonstrated that item-level analyses provide deeper insight into how depressive symptoms, anxiety, and perceived stress interact among university students, which is relevant for future research and applications in mental health. (Less)
Abstract (Swedish)
Depression, ångest och stress är närliggande former av psykisk ohälsa som ofta uppvisar betydande symtomatisk överlappning. Föreliggande studie undersökte hur enskilda frågor inom självskattningsinstrumenten PHQ-9, GAD-7 och PSS-10 relaterade till respektive skala samt till varandra bland universitetsstudenter,
med fokus på vilka frågor som framstod som mest centrala och indikativa för variation i depressions-, ångest- och stressnivåer. Analysen baserades på stora, öppet tillgängliga dataset omfattande över 50 000 universitetsstudenter från flera länder. Analyserna inkluderade deskriptiv statistik, item–total-korrelationer,
beräkning av intern konsistens (Cronbachs α), konfirmatorisk faktoranalys, regulariserade LASSO-regressioner samt... (More)
Depression, ångest och stress är närliggande former av psykisk ohälsa som ofta uppvisar betydande symtomatisk överlappning. Föreliggande studie undersökte hur enskilda frågor inom självskattningsinstrumenten PHQ-9, GAD-7 och PSS-10 relaterade till respektive skala samt till varandra bland universitetsstudenter,
med fokus på vilka frågor som framstod som mest centrala och indikativa för variation i depressions-, ångest- och stressnivåer. Analysen baserades på stora, öppet tillgängliga dataset omfattande över 50 000 universitetsstudenter från flera länder. Analyserna inkluderade deskriptiv statistik, item–total-korrelationer,
beräkning av intern konsistens (Cronbachs α), konfirmatorisk faktoranalys, regulariserade LASSO-regressioner samt nätverksanalyser på item-nivå. Resultaten visade att samtliga instrument uppvisade tydliga frågespecifika mönster, där vissa symtom hade starkare samband med övriga frågor än andra. Nätverksanalyserna indikerade att affektiva och kognitiva symtom ofta intog mer centrala positioner i nätverken, medan somatiska symtom generellt var mer perifera. Sammantaget visade studien att analys på item-nivå ger fördjupad kunskap om hur depressiva symtom, ångest och upplevd stress samverkar hos universitetsstudenter, vilket är relevant för framtida forskning och tillämpning inom psykisk hälsa. (Less)
Please use this url to cite or link to this publication:
author
Bernborg, Emma LU and Watts, Elliot LU
supervisor
organization
course
PSYK12 20252
year
type
M2 - Bachelor Degree
subject
keywords
PHQ-9, GAD-7, PSS-10, Network analysis, LASSO-regression, Nätverksanalys
language
Swedish
id
9220998
date added to LUP
2026-01-28 10:15:17
date last changed
2026-01-28 10:15:17
@misc{9220998,
  abstract     = {{Depression, anxiety, and stress are closely related forms of psychological distress that often exhibit substantial symptomatic overlap. The present study examined how individual items within the self-report instruments PHQ-9, GAD7, and PSS-10 were related to their respective scales as well as to each other
among university students, with a focus on which items appeared most central and indicative of variation in levels of depression, anxiety, and stress. The analyses were based on large, openly available datasets comprising over 50,000 university students from multiple countries. The analyses included descriptive statistics, item–total correlations, estimation of internal consistency (Cronbach’s α), confirmatory factor analysis, regularized LASSO regressions, and network analyses at the item level. The results showed that all instruments exhibited clear item-specific patterns, with certain symptoms showing stronger associations
with other items than others. Network analyses indicated that affective and cognitive symptoms often occupied more central positions in the networks, whereas somatic symptoms were generally more peripheral. Overall, the study demonstrated that item-level analyses provide deeper insight into how depressive symptoms, anxiety, and perceived stress interact among university students, which is relevant for future research and applications in mental health.}},
  author       = {{Bernborg, Emma and Watts, Elliot}},
  language     = {{swe}},
  note         = {{Student Paper}},
  title        = {{Depression, ångest och stress hos universitetsstudenter: en tvärsnittsstudie med nätverks- och LASSO-baserad analys av PHQ-9, GAD-7 och PSS-10}},
  year         = {{2026}},
}