Loneliness Scale

1. Abstract

The De Jong Gierveld Loneliness Scale (originally developed in Dutch as the Eenzaamheidsschaal; de Jong-Gierveld & Kamphuis, 1985; de Jong Gierveld & Van Tilburg, 1999) is an internationally recognized, multidimensional psychometric instrument designed to assess subjective experiences of loneliness among adults and older adult populations. Developed in response to the limitations of single-item global indicators and unidimensional scales, the instrument is grounded in a cognitive-discrepancy theoretical framework that conceptualizes loneliness not merely as objective social isolation, but as an evaluated deficit between an individual’s actual interpersonal network and their subjective relational standards and desires. The standard instrument comprises 11 self-report items divided into two interrelated yet distinct subscales: Emotional Loneliness (6 negatively formulated items capturing the absence of intimate, deeply affectionate attachments and feelings of emptiness or abandonment) and Social Loneliness (5 positively formulated items reflecting the perceived lack of a broader, supportive social network, circle of friends, and community embeddedness).

Items are traditionally administered using a 3-point categorical response format (“Yes”, “More or less”, “No”) or a standardized 5-point Likert scale, which are subsequently dichotomized into an aggregate index ranging from 0 (absence of loneliness) to 11 (severe loneliness), alongside distinct continuous or categorical subscale scores. Over four decades of empirical evaluation across diverse cross-national cohorts—including the Generations and Gender Programme (GGP) and extensive European longitudinal aging panels—have established that the scale demonstrates robust psychometric properties. Internal consistency estimates typically range from Cronbach’s α = .80 to .88 for the total scale, with satisfactory reliability observed across both the emotional (α ≈ .74–.82) and social (α ≈ .76–.84) dimensions. Confirmatory factor analyses, item response theory (Mokken scale analysis), and structural equation models substantiate a robust two-factor oblique structure that remains invariant across gender and age strata. The scale exhibits high convergent validity with depressive symptomatology, perceived stress, and poor self-rated health, as well as divergent validity with objective social network size and living arrangements.

2. Keywords

Loneliness Scale, De Jong Gierveld Loneliness Scale, Eenzaamheidsschaal, emotional loneliness, social loneliness, cognitive discrepancy theory, psychometrics, social isolation, geriatric assessment, Mokken scale analysis

3. Authors

The scale was developed and psychometrically refined by two prominent Dutch sociologists and social demographers at the Vrije Universiteit Amsterdam and the Netherlands Interdisciplinary Demographic Institute (NIDI):

• Jenny de Jong-Gierveld, Ph.D. — Professor Emeritus of Social Demography and Social Gerontology at Vrije Universiteit Amsterdam and Senior Research Fellow at the Netherlands Interdisciplinary Demographic Institute (NIDI-KNAW), The Hague, Netherlands. Renowned for her foundational work on social relationships, loneliness, living arrangements, and aging.
• Frits H. Kamphuis, Ph.D. — Sociologist and psychometrician who collaborated with de Jong-Gierveld during the early 1980s at the Department of Sociology and Social Research Methodology, Vrije Universiteit Amsterdam, contributing specifically to the latent trait modeling and item calibration of the original instrument.
• Theo G. van Tilburg, Ph.D. (Key Contributor & Co-developer of the 6-item short form) — Professor of Sociology and Social Gerontology, Department of Sociology, Vrije Universiteit Amsterdam, whose psychometric collaborations advanced cross-national validation and latent structure refinement.

Institutional Contact Information:
Netherlands Interdisciplinary Demographic Institute (NIDI)
PO Box 11650, NL-2502 AR The Hague, The Netherlands
Website: https://nidi.nl

4. Purpose

The primary purpose of the De Jong Gierveld Loneliness Scale is to provide a theoretically driven, empirically rigorous assessment of loneliness that differentiates between qualitative relational deprivation and broader network deficiencies. Historically, epidemiological and social surveys evaluated loneliness through single-item questions (e.g., “Do you ever feel lonely?”), which suffered from substantial social desirability bias, cultural stigma, and an inability to disentangle the distinct psychological facets of interpersonal isolation. Other historical instruments, such as the initial UCLA Loneliness Scale, operated largely under a unidimensional paradigm. De Jong-Gierveld and Kamphuis sought to develop an instrument that could overcome these limitations by using non-pejorative, nuanced phrasing that avoids the direct word “lonely” in its item stems, thus reducing defensive underreporting while offering detailed clinical and epidemiological insight.

In clinical, geriatric, and epidemiological settings, the scale serves several critical functions:

• Differential Diagnosis of Relational Deficits: The scale distinguishes between individuals who suffer from the absence of an intimate romantic or personal attachment figure (emotional loneliness) versus those who lack an extended social matrix of peers, colleagues, or community members (social loneliness). This differentiation is critical for interventions; introducing a senior center group activity may alleviate social loneliness, whereas it rarely resolves emotional loneliness stemming from spousal bereavement.
• Monitoring Public Health and Gerontological Interventions: Social prescribing initiatives, community engagement programs, and psychotherapy protocols utilize the scale as a pre- and post-intervention outcome benchmark to quantify therapeutic efficacy across distinct relational domains.
• Population-Based Health Surveillance: Because loneliness is an established, independent risk factor for cardiovascular disease, neurocognitive decline, major depressive disorder, and all-cause mortality, public health researchers deploy the instrument to monitor demographic shifts, assess vulnerable cohorts (e.g., widowed elders, marginalized migrant groups), and inform policy regarding social cohesion.
• Longitudinal Research: The scale enables developmental psychologists and sociologists to track how life-course transitions—such as retirement, divorce, institutionalization, and physical impairment—selectively degrade specific facets of interpersonal security over time.

5. Psychological Construct

The core psychological construct measured by the De Jong Gierveld Loneliness Scale is subjective loneliness, defined as a cognitive and emotional state of distress experienced when an individual perceives an unacceptable discrepancy between the quality, quantity, and intimacy of their actual social network and their internal relational aspirations. The construct is inherently multidimensional, operationalized across two correlated yet separate domains:

Emotional Loneliness

Emotional loneliness arises from the perceived absence or rupture of a primary, close, intimate attachment figure. It represents a profound sense of inner emptiness, abandonment, and the absence of someone with whom one can share unconditional vulnerability, deep affection, and mutual commitment. Emotional loneliness is most acutely triggered by structural losses such as the death of a spouse or life partner, marital dissolution, or geographic separation from key attachment figures.

Within the scale, emotional loneliness is captured by 6 negatively worded items (Items 2, 3, 5, 6, 9, and 10) that measure internal feelings of deprivation:

• Deficit in Intimate Confidants: Exemplified by Item 2 (“I miss having a really close friend”), evaluating the profound lack of a dedicated attachment partner.
• Existential Emptiness and Dejection: Reflected in Item 3 (“I experience a general sense of emptiness”) and Item 10 (“I often feel rejected”), capturing the dysphoric and alienating psychological consequences of lacking deeply validating relationships.
• Relational Yearning: Evidenced by Item 5 (“I miss the pleasure of the company of others”), Item 6 (“I find my circle of friends and acquaintances too limited”), and Item 9 (“I miss having people around”), which record the emotional pain of longing for deeper companionship.

Social Loneliness

In contrast, social loneliness stems from the perceived absence of a broader social network or community of peers, friends, neighbors, or colleagues who provide a sense of belonging, companionship, shared identity, and practical instrumental assistance. An individual may possess a deeply loving partner (thus not experiencing emotional loneliness) but feel socially isolated if they lack broader community integration, casual friendships, and group affiliations.

Within the scale, social loneliness is assessed via 5 positively worded items (Items 1, 4, 7, 8, and 11) that represent social integration and instrumental availability, which are scored as lonely when the respondent indicates their absence:

• Instrumental and Problem-Solving Support: Reflected in Item 1 (“There is always someone I can talk to about my day-to-day problems”) and Item 11 (“I can call on my friends whenever I need them”).
• Trust and Reliance: Captured by Item 4 (“There are plenty of people I can lean on when I have problems”) and Item 7 (“There are many people I can trust completely”), assessing psychological safety within an extended network.
• Perceived Network Embeddedness: Measured by Item 8 (“There are enough people I feel close to”), determining whether an individual feels anchored in a viable social circle.

6. Theoretical Framework

The architectural foundation of the De Jong Gierveld Loneliness Scale is built upon two major theoretical paradigms: Weiss’s Relational Provisions Theory (Weiss, 1973) and the Cognitive Discrepancy Model of Loneliness (Peplau & Perlman, 1982; de Jong-Gierveld, 1987).

Weiss’s Typology of Relational Provisions

Robert S. Weiss (1973) transformed loneliness research by positing that human beings require multiple distinct interpersonal provisions that cannot be substituted for one another. Weiss identified two core dimensions of relational deficiency:

1. The Loneliness of Emotional Isolation: Resulting from the absence of an attachment figure providing emotional integration, felt security, and intimate sharing (traditionally supplied by a spouse, romantic partner, or intimate confidant).
2. The Loneliness of Social Isolation: Resulting from the absence of an engaging social network that offers social integration, shared concerns, recreational companionship, and a sense of collective belonging.

De Jong-Gierveld and Kamphuis operationalized Weiss’s theoretical distinction into empirical measurement, demonstrating that emotional and social loneliness show differential associations with demographic characteristics, psychiatric morbidities, and life events.

The Cognitive Discrepancy Framework

Supplementing Weiss’s categorical distinction, de Jong-Gierveld integrated the cognitive perspective developed by Letitia Anne Peplau and Daniel Perlman. In this framework, loneliness is conceptualized as an affective cognitive reaction that occurs when an individual perceives an unacceptable gap between their perceived existing relationships (what they currently have) and their internal relational standard (what they desire, expect, or feel entitled to). This cognitive formulation explains why two individuals with identical social networks may experience vastly different levels of loneliness: an introverted individual with few contacts may experience no loneliness because their network meets their standard, whereas an extroverted individual surrounded by dozens of acquaintances may experience severe loneliness if their standard calls for intimate, unconditional bonding.

Importantly, the scale incorporates Mokken Latent Trait Theory (Mokken, 1971), a non-parametric item response theory (IRT) framework. Rather than assuming equal item difficulties, the 11 items were calibrated to represent cumulative, hierarchically ordered levels of perceived relational deprivation along a single latent continuum, ensuring high structural precision across varying degrees of loneliness severity.

7. Validity

The psychometric validity of the De Jong Gierveld Loneliness Scale has been extensively documented in clinical samples, general community surveys, and cross-national epidemiological projects across North America, Europe, and Asia.

Construct and Structural Validity

Structural equation modeling and multi-group confirmatory factor analysis (CFA) consistently confirm that a two-factor model (Emotional Loneliness and Social Loneliness) yields superior goodness-of-fit indices compared to a unidimensional model across diverse cultures and languages (e.g., comparative fit index [CFI] > .95; root mean square error of approximation [RMSEA] < .06). While emotional and social loneliness are positively correlated (inter-factor correlations typically range between r = .35 and .55), they maintain distinct statistical identities, validating their structural independence.

Convergent Validity

The scale demonstrates strong convergent validity when compared against other established loneliness and mental health instruments:

• Total scale scores correlate strongly with the UCLA Loneliness Scale (correlations typically ranging from r = .70 to .82).
• The instrument shows substantial positive correlations with depression measures, including the Center for Epidemiologic Studies Depression Scale (CES-D) (r = .50 to .62) and the Geriatric Depression Scale (GDS) (r = .48 to .58).
• Moderate positive correlations are observed with neuroticism, psychological distress, perceived stress, and perceived physical health deficits.

Divergent and Discriminant Validity

The scale successfully differentiates subjective loneliness from objective social isolation:

• Correlations with objective network metrics (such as network size, frequency of face-to-face contacts, and marital or cohabitation status) are modest (typically r = −.20 to −.35 for social loneliness, and r = −.25 to −.40 for emotional loneliness), empirically confirming that objective isolation and subjective loneliness are distinct phenomena.
• The emotional loneliness subscale discriminates effectively between married or partnered individuals and those who are widowed, divorced, or single, exhibiting significantly higher scores among the latter. Conversely, the social loneliness subscale discriminates between individuals based on community involvement, volunteer participation, and active friendship circles regardless of marital status.

Predictive and Criterion Validity

Longitudinal prospective cohort studies have demonstrated that elevated baseline scores on the scale predict significant downstream clinical endpoints, including increased incidence of major cardiovascular events, functional mobility limitations, cognitive impairment and dementia incidence, nursing home admission, and all-cause mortality over 5- to 10-year follow-up windows, even after controlling for baseline sociodemographic and medical confounders.

8. Reliability

The reliability of the De Jong Gierveld Loneliness Scale has been demonstrated using classical test theory, item response theory, and longitudinal stability metrics:

Internal Consistency

Extensive validation studies across multiple European cohorts (such as the NESTOR study and the Longitudinal Aging Study Amsterdam [LASA]) report high internal consistency for the overall 11-item scale, with Cronbach’s α coefficients consistently ranging from .80 to .88. Reliability coefficients for the specific subscales are robust across age categories:

• Emotional Loneliness Subscale (6 items): Cronbach’s α typically ranges from .74 to .82.
• Social Loneliness Subscale (5 items): Cronbach’s α typically ranges from .76 to .84.

Mokken scale analysis confirms high homogeneity across the item set, with Loevinger’s H coefficients generally exceeding .40 to .50, satisfying the criteria for strong, cumulative, unidimensional subscale hierarchies under non-parametric IRT assumptions.

Test-Retest Stability

Test-retest reliability reflects the theoretical conceptualization of loneliness as a subjective state with an underlying dispositional (trait-like) baseline:

• Over short intervals (2 to 4 weeks), stability coefficients remain high (r = .80 to .87), indicating excellent measurement consistency.
• Over extended intervals (1 to 3 years in longitudinal gerontological studies), test-retest correlations hover between r = .55 and .68, reflecting both chronic baseline vulnerabilities and meaningful psychological changes driven by life events such as partner bereavement, relocation, or health declines.

9. Factor Analysis

The structural dimensionality of the 11-item scale has been investigated through Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and Item Response Theory across several decades.

Confirmatory Factor Analysis (CFA) Fit Statistics

While early psychometric evaluations debated whether the scale was unidimensional (with item formulation method effects driving a two-factor split), extensive multi-national CFA studies have definitively affirmed an oblique two-factor structural model separating Emotional Loneliness from Social Loneliness. Exemplary model fit parameters reported across large population datasets (e.g., van Baarsen et al., 2001; Grygiel et al., 2013) include:

• Comparative Fit Index (CFI): .96 to .98 (exceeding the standard .95 threshold for excellent model fit)
• Tucker-Lewis Index (TLI): .95 to .97
• Root Mean Square Error of Approximation (RMSEA): .038 to .052 (90% CI [.030, .060])
• Standardized Root Mean Square Residual (SRMR): .035 to .048

Factor Loadings and Factor Composition

Measurement Invariance

Cross-national invariance testing (e.g., across French, German, Polish, Italian, and Dutch populations within the Generations and Gender Programme) demonstrates full configural and metric invariance, and partial to full scalar invariance. This confirms that the underlying latent constructs are equivalent across diverse linguistic groups, age strata (young adults, middle-aged adults, oldest-old), and genders, enabling comparative demographic research.

10. Instrument / Measurement Tool

• Instrument Name: De Jong Gierveld Loneliness Scale (Eenzaamheidsschaal)
• Authors: Jenny de Jong-Gierveld & Frits H. Kamphuis (1985)
• Target Population: Adults (18+), middle-aged, and older adult populations.
• Administration Format: Self-administered paper-and-pencil questionnaire, computer-assisted personal interviewing (CAPI), telephone interview, or web-based survey.
• Administration Time: Approximately 3 to 5 minutes.
• Total Item Count: 11 items.
• Subscales:
  • Emotional Loneliness: 6 items (Items 2, 3, 5, 6, 9, 10) — all negatively formulated.
  • Social Loneliness: 5 items (Items 1, 4, 7, 8, 11) — all positively formulated.
• Authentic Response Scale: 3-point response format: Yes / More or less / No (also commonly administered on a 5-point Likert scale: 1=Strongly agree, 2=Agree, 3=More or less, 4=Disagree, 5=Strongly disagree).
• Standard Dichotomous Scoring Rules:
  • Negatively Formulated Items (Emotional Loneliness: 2, 3, 5, 6, 9, 10): The response reflecting a deficit scores 1, and the absence of a deficit scores 0.
    • “Yes” = 1
    • “More or less” = 1
    • “No” = 0

    (If using a 5-point Likert scale: Strongly agree = 1, Agree = 1, More or less = 1, Disagree = 0, Strongly disagree = 0).

  • Positively Formulated Items (Social Loneliness: 1, 4, 7, 8, 11): The response reflecting an absence of social resources scores 1, and positive endorsement scores 0.
    • “No” = 1
    • “More or less” = 1
    • “Yes” = 0

    (If using a 5-point Likert scale: Strongly agree = 0, Agree = 0, More or less = 1, Disagree = 1, Strongly disagree = 1).
• Total Score Calculation & Clinical Cut-offs:
  • Sum of all 11 dichotomized items yields an overall score ranging from 0 to 11.
  • Score 0 – 2: Not lonely (socially and emotionally integrated).
  • Score 3 – 8: Moderate loneliness.
  • Score 9 – 10: Severe loneliness.
  • Score 11: Very severe / extreme loneliness.
  • Subscale Scores: Emotional Loneliness ranges from 0 to 6; Social Loneliness ranges from 0 to 5.

11. Permissions & Fee and Test Year

The original Dutch 11-item Loneliness Scale was published in 1985 by Jenny de Jong-Gierveld and Frits H. Kamphuis in the Journal of Personality and Social Psychology. In 1999 and 2006, de Jong Gierveld and Theo van Tilburg published extensive methodological updates and introduced an abbreviated 6-item version designed specifically for large-scale survey contexts.

Licensing and Accessibility:
The De Jong Gierveld Loneliness Scale is in the public domain for academic, scientific, and non-commercial clinical research purposes. Researchers and healthcare practitioners may utilize the scale without royalty fees, provided that appropriate academic attribution and citation of the foundational development papers are given. Commercial organizations, survey consultancies, or commercial digital health applications seeking to integrate the scale should consult the primary copyright holders and institutional affiliates at the Netherlands Interdisciplinary Demographic Institute (NIDI-KNAW).

12. References

• de Jong-Gierveld, J., & Kamphuis, F. (1985). The development of a Rasch-type loneliness scale. Journal of Personality and Social Psychology, 49(2), 505–517. https://doi.org/10.1037/0022-3514.49.2.505
• de Jong-Gierveld, J. (1987). Developing and testing a model of loneliness. Journal of Personality and Social Psychology, 53(1), 119–128. https://doi.org/10.1037/0022-3514.53.1.119
• de Jong Gierveld, J., & Van Tilburg, T. (1999). Manual of the loneliness scale. Department of Social Research Methodology, Vrije Universiteit Amsterdam, 1–26.
• de Jong Gierveld, J., & Van Tilburg, T. (2006). A 6-item scale for overall, emotional, and social loneliness: Confirmatory tests on survey data. Research on Aging, 28(5), 582–598. https://doi.org/10.1177/0164027506289723
• Grygiel, P., Humenny, G., Rebisz, S., Switaj, P., & Sikorska-Grygiel, J. (2013). Validating the Polish adaptation of the 11-item De Jong Gierveld Loneliness Scale. European Journal of Psychological Assessment, 29(2), 129–136. https://doi.org/10.1027/1015-5759/a000130
• Mokken, R. J. (1971). A Theory and Procedure of Scale Analysis: With Applications in Political Research. De Gruyter Mouton. https://doi.org/10.1515/9783110813203
• Peplau, L. A., & Perlman, D. (Eds.). (1982). Loneliness: A Sourcebook of Current Theory, Research and Therapy. John Wiley & Sons.
• van Baarsen, B., Snijders, T. A., Smit, J. H., & van Duijn, M. A. (2001). Lonely but not alone: Emotional isolation and social isolation as two distinct dimensions in loneliness among older adults. Social Indicators Research, 53(2), 119–152. https://doi.org/10.1023/A:1007137126154
• Weiss, R. S. (1973). Loneliness: The Experience of Emotional and Social Isolation. MIT Press.

13. Items of the Scale