Introduction

There is evidence of the positive effects of participation and social support among older adults (OA) in groups and organizations beyond the family on their health, well-being, and quality of life (1–4). Social participation (SP) influences differences in life expectancy and healthy life years among OA in Chile and other countries (5). Those who belong to older adult clubs show better health and well-being, with differences by sex (4). According to Fernández et al. (6), social support from family, friends, and community participation promotes healthy and active aging. Social participation, a polysemic concept, includes multiple social behaviors and is defined as “involvement in activities that provide interaction with others in society or community” (7 p2148).

Systematic reviews on the effects of SP on health (1) show that, with variations depending on the activity type, all forms yield physical and mental benefits, delaying deterioration in OA. SP is associated with lower physical and cognitive decline, better health and daily functioning, lower mortality, and higher quality of life. Fu et al. (2) found that SP increases physical activity and protects against the effects of frailty. SP and social support, along with instrumental activity, frailty, and loneliness, have been central research topics regarding OA during the last five years (2). Having instruments that adequately measure social support networks in OA is of high diagnostic relevance because it allows predicting risks of unwanted loneliness—different from voluntary disengagement—and its consequences, including distress and health risks.

A brief scale with adequate psychometric properties and widely used internationally, according to Granero et al. (8), is the Lubben Social Network Scale (LSNS-6), by Lubben et al. (9). Its original 18 items have been reduced to 6 in Japan (10), several European cities (9), Portugal (11), Korea (12), Mongolia (13), China (14), Argentina (8), and Puerto Rico (15), consistently confirming its bidimensionality, reliability, and validity. Nunnally (16) recommends reliability values of 0.70 or higher – but not much higher than 0.80 – for basic research. The Japanese version (10) reported Cronbach’s alpha of 0.82, test-retest of 0.92 (p < .001), and scale scores increased with the number of family members and decreased among OA at suicide risk. The Chinese version (14), applied to adults aged 60–90 (mean 70.8), reported Cronbach’s alpha of 0.83, with 0.90 for the Family dimension and 0.95 for Friends, explaining 86.8% of variance. High Lubben-6 scores and Family subscale scores were associated with lower suicide risk (ORs < 1, p < .05). In Mongolia, among 650 adults aged 55–75+, Myagmarjav et al. (13) reported Cronbach’s alpha of 0.86 (0.82 Family; 0.87 Friends), and test-retest of 0.91. Lubben-6 correlated negatively with social disengagement (r = –0.52, p < .001), perceived isolation (r = –0.54, p < 0.001), and depression (r = –0.27, p<0.001).

Among residents of elder-care facilities in three European cities (9)—Hamburg (n=1964), Solothurn (n=2870), and London (n=2598)—with a mean age of 74, internal consistency was 0.73, 0.74, and 0.74 respectively. In Portugal, among 207 residents aged 65–96 (mean 77.2), with up to 4 years of schooling (89%), Cronbach’s alpha was 0.798 (0.756 Family; 0.732 Friends). Two factors explained 67.8% of variance. No sex differences were found for Family; men scored higher on Friends. Comparing age groups (65–74, 75–84, 85+), older adults scored lower on Friends.

In Hispanic America, the Puerto Rican version (15), applied to 154 OA aged 60–96 (mean 73.8, 61.7% women), reported Cronbach’s alpha of 0.74; Family 0.82; Friends 0.67. Low scores correlated significantly and moderately with loneliness (r = –0.33 p < 0.001), and a larger friend network was associated with better self-rated health. The Argentine version (8), in 150 OA (mean 76.3; 54% men), showed satisfactory composite reliability for both factors (0.83 and 0.91), with extracted mean variance of 0.62 and 0.75. Factor analysis confirmed a bifactorial model with adequate fit values (RMSEA 0.085, NFI 0.97), with standardized loadings above 0.6 or 0.7. The Pearson correlation between Lubben-6 and social-health worker evaluation was 0.665 (p < 0.001).

The objective of this study is to adapt and advance validation of the Argentine Lubben-6 Scale (8) for Chile, considering its psychometric strength and linguistic, cultural, and geographic proximity.

Methodology

Design: Cross-sectional psychometric study with urban samples of OA from four Chilean cities: one in the north, two in the center, and one in the south. Data collection occurred between November 2023 and March 2024.

Sample: Convenience sample of men aged 65+ and women aged 60+, totaling 346 participants (76.8% women), from Antofagasta (north), Rancagua and Talca (center), and Punta Arenas (south). Mean age was 71.7 (SD=6.85) (Table 1). Exclusion criteria: illiteracy or inability to independently understand and answer the survey. Ethics approval was granted by the Scientific Ethics Committee of the University of Talca (Folio 03-2023).

Table 1. Descriptive characteristics of the sample of older people.

Measures: Survey including: (i) Lubben-6 Social Network Scale. (ii) Hennessy self-rated health scale (17). (iii) EFPA happiness scale (18). (iv) 14 items on social engagement / disengagement, autonomy, sexual satisfaction, family and friend satisfaction, leisure, work, fear of death, sense of peace, and happiness sourcesProcedure: Local coinvestigators recruited participants and administered surveys. Analyses included descriptives, reliability, confirmatory factor analysis, and multigroup invariance testing. Reliabilitythreshold: .70–.90.

Data Analysis: Descriptives and confirmatory factor analysis for 1- and 2-factor models. Fit indices: Chi-square, CFI ≥ .90, TLI ≥ .90, RMSEA < .08, NFI ≥ .90. Criterion validity: correlations with engagement, disengagement, and EFPA. Software: SPSS v25, AMOS v23.

Results

Table 2 shows the mean, standard deviation, skewness, kurtosis, and mode values obtained for the items of the Lubben-6 scale. All items meet the normality criteria for skewness and kurtosis, which range between -2 and 2 (21).

Analysis of the psychometric properties of the Lubben-6 scale

Table 2. Means, standard deviations, skewness, kurtosis, mode, for the Lubben-6 items.

Cronbach’s alpha for the total Lubben-6 scale is 0.766, broken down into 0.710 for the Family dimension and 0.776 for Friends. The corresponding Omega values are 0.776 (total scale), 0.716 (family dimension), and 0.781 (friends’ dimension) (Table 3).

Table 3. Cronbach’s Alpha coefficients for the Lubben-6 Scale and its subscales

The model diagram and factor loadings for each Lubben-6 item are presented below (Figure 1).

Construct validity. The confirmatory factor análisis (22) for the Lubben-6 reproduced its original two-dimensional structure (χ² = 46.23, df = 8, p < .001, RMSEA = .118, CFI = .934, TLI = .876, NFI = .922).

A one-factor model showed a poor fit (χ² = 166.69, df = 9, p < .001, RMSEA = .225, CFI = .727, TLI = .546, NFI = .719).

Figure 1: Model for the Lubben-6 Red Social Support Scale and factor loadings per item.
Figure 1: Model for the Lubben-6 Red Social Support Scale and factor loadings per item.
Source: Prepared by the authors

To test the measurement invariance of the Lubben-6 for the male and female subgroups, a multi-group confirmatory factor analysis was performed. The metric or weak invariance test, where factor loadings (λ) are restricted so that they are equal between sex groups, showed a good fit according to the criteria established by Cheung et al. (22). When comparing M1 with M2, no significant changes were observed in the CFI, RMSEA, or χ² indices. For the scalar invariance model, where factor loadings (λ) and intercepts (τ) are restricted, the CFI and RMSEA indices are within tolerable ranges, but the χ² index shows differences in the measures between sex groups, indicating a partial fit (Table 4).

Table 4. Results of metric, scalar and strict invariance analysis, by sex, for the Lubben-6.

Criterion validity. Four procedures were used to analyze and establish this type of validity. First, bivariate correlations-expected to be positive-were performed between the Lubben-6, its dimensions, the total EFPA, and the dimensions of the EFPA (Table 5).

Table 5. Spearman bivariate correlations between the Lubben-6, EFPA scales and their respective dimensions (n=346).

A positive and significant correlation was confirmed between the Lubben-6 and the EFPA happiness scale (.246**). The Family Support dimension of the Lubben-6 and the Family Happiness dimension of the EFPA also correlated positively and significantly (.237**). The Friends Support dimension, in addition to correlating positively and significantly with the total EFPA scale (.232**), also correlated with its Balance and Harmony dimension (.235**). Furthermore, the Friends Support dimension of the Lubben-6 correlated positively with the EFPA’s Achievement-Optimism Happiness dimension, which includes individual and collective achievements (.154*).

Secondly, the Friends factor of the Lubben-6 was correlated with the 14 survey items related to the decision to have more or less of a social life, to seek and have more friends, and the level of satisfaction with them—items 46 (I decided to seek and have more social life or friends) and 53 (I feel satisfied or happy with my friends)—and, as expected, a positive and significant correlation was found with these two items (r = .193, p = .001; r = .346, p = .001, respectively). The Family Support subscale, however, was not correlated with any of these items (with item 46, r = .066, p = .223 and r = .073, p = .177, respectively).

Third, the Lubben-6 was correlated with the survey items related to social disengagement—items 43, 44, and 45— and, as expected, significant negative correlations were found (r = -.177, p = .001; r = -.127, p = .028; r = -.237, p = .001, respectively).

Fourth, no significant differences in mean scores were observed by sex for the total Lubben-6 scale (M men = 2.34, SD = .91 and M women = 2.56, SD = .87, t = -1.93, p = .054). Nor for the Family Support subscale (t = -.31, p = .75), while for the Friends Support subscale statistically significant differences are reported (M.men = 1.75, SD = 1.26 and M.women = 2.15, SD = 1.15, t = -2.63, p = .001), in favor of the latter.

Discussion

The reliability of the Lubben-6 is 0.766 for Cronbach’s Alpha and Omega coefficients, and those for each of its two dimensions are acceptable, with values greater than .70 for both. These values are very similar to those obtained in the samples from the three Central European cities, Portugal, and Latin America, compared to those reported for Puerto Rico. Furthermore, a good fit of the items to the bidimensionality of the Lubben-6 is observed. Although the RMSEA fit index does not exceed the cutoff point proposed by the literature, most fit indices indicate adequate construct validity of the instrument. The literature has shown that the RMSEA fit index tends to show erroneous values toward an increase when the models are complex and have few degrees of freedom, which could be affecting this instrument (23). Even so, the results replicate those of previous validation processes, where the instrument’s two-factor structure shows a much better fit than a one-factor structure (8). Additionally, the correlation between its two dimensions is .39, demonstrating that they are two distinct factors, with the one related to Friends appearing as a greater influence or predictor of social bonding than the one related to Family.

At a descriptive level, the trends related to having close friends are minimal (1), which is consistent with recent national data, and broader age ranges, showing the risk of loneliness for Chileans aged 18 and over, where 15% report not having a close friend (while in the USA the percentage was only 8%) (24).

Although the structure of this instrument remains invariant according to sex, this is only partially true, given that the requirements for scalar or strict invariance are not fully met (25). However, the specialized literature has concluded that strict invariance models are excessively restrictive (26). Thus, the results indicate that the instrument scores can be compared between the tested groups of different sexes.

Finally, very clear support for criterion validity is obtained, in line with what has been reported in previous studies, with adequate correlations in both direction and statistical significance. The Lubben-6 correlates positively with the happiness of older adults. Correlations between specific variables also show the expected positive direction, as is the case with the Family dimension of both instruments (Lubben-6 and EFPA), and measures of perceived family support. Positive correlations were also found between older adults’ decisions to increase their social life and the Friends dimension, and negative correlations with decisions regarding Social Disengagement.

Among the weaknesses of this study is the unequal number of participants by sex in the sample. Although this adequately represents the typical distribution of men and women in older adult clubs, church groups, and similar settings in the country, it hinders certain statistical analyses and their results, e.g., the invariance analysis. While this analysis could be performed, its results would be more significant with a more balanced sample. Also, although the sample size meets the minimum requirements for a validation study, a larger number of participants would be desirable. Finally, although unlikely, it is possible that some Spanish terms used in the adapted Argentinian version used here may not be exactly the same as their usage in Chile.

Conclusions

The Lubben-6 shows adequate psychometric indices in samples of the Chilean population. Its Cronbach’s alpha and omega reliability are 0.766, and by factor, it reaches a Cronbach’s alpha of 0.71 for family (0.716 omega) and a Cronbach’s alpha of 0.776 for friends (0.781 omega). Its two-factor structural and criterion validity is confirmed by positive associations with measures of social integration and well-being, and negative associations with social disengagement.

Thus, the Lubben-6 Social Support Network scale can be used reliably and validly in the Chilean elderly population.

Conflicts of interest

The authors declare that they have no conflicts of interest

Funding

This study was funded by the Interuniversity Center for Healthy Aging (CIES) under project approval number CIES-003.

Ethical considerations

The study was approved by the Scientific Ethics Committee of the University of Talca (Chile) CEC, under Folio No. 03-2023.

Confidentiality: Informed consent was obtained, and the information obtained is anonymous and remains stored and protected according to the corresponding protocol.

Appendix 1. Items to measure social engagement/disengagement, sexual satisfaction, family life, partner relationships, friendships, leisure and work life, autonomy in decisions regarding one’s own assets, fear of death, sense of peace, and sources of happiness. Choose by marking an X in the box of your choice, ranging from 1 = Strongly Disagree to 5 = Strongly Agree

References

1. Stav WB, Hallenen T, Lane J, Arbesman M. Systematic review of occupational engagement and health outcomes among community-dwelling older adults. Am J Occup Ther [Internet]. 2012;66(3):301-10. doi: https://doi.org/10.5014/ajot.2012.003707

2. Fu J, Jiang Z, Hong Y, Liu S, Kong D, Zhong Z., et al. Global scientific research on social participation of older people from 2000 to 2019: Abibliometric analysis. Int J Older People Nurs [Internet]. 2021;16(1):e12349. doi: https://doi.org/10.1111/opn.12349 f

3. Douglas H, Georgiou A, Westbrook J. Social participation as an indicator of successful aging: an overview of concepts and their associations with health. Aust Health Rev [Internet]. 2017;41(4):455-62. doi: https://doi.org/10.1071/ah16038

4. Wanchai A, Phrompayak D. Social participation types and benefits on health outcomes for elder people: a systematic review. Ageing Int [Internet]. 2019;44(3):223-233. doi: https://doi.org/10.1007/s12126-018-9338-6

5. Rueda-Salazar S, Spijker J, Devolder D, Albala C. The contribution of social participation to differences in life expectancy and healthy years among the older population: A comparison between Chile, Costa Rica and Spain. PLoS ONE [Internet]. 2021;16(3):e0248179. doi: https://doi.org/10.1371/journal.pone.0248179

6. Fernández F, Nazar G, Alcover M. Modelo de envejecimiento activo: causas, indicadores y predictores en adultos mayores en Chile. Acción Psicol [Internet]. 2018;15(2):109-128. doi: https://doi.org/10.5944/ap.15.2.22903

7. Levasseur M, Richard L, Gauvin L, Raymond É. Inventory and analysis of definitions of social participation found in the aging literature: Proposed taxonomy of social activities. Soc Sci Med [Internet]. 2010;71(12):2141-2149. doi: https://doi.org/10.1016/j.socscimed.2010.09.041

8. Granero M, Perman G, Vazquez-Peña F, Barbaro CA, Zozaya ME, Martínez-Infantino VS, et al. Validación de la versión en español de la escala de red social Lubben-6. Rev Fac Cien Med Univ Nac Cordoba [Internet]. 2020;77(4):296-300. doi: https://doi.org/10.31053/1853.0605.v77.n4.28775

9. Lubben J, Blozik E, Gillmann G, Iliffe S, von Renteln-Kruse W, Beck JC, et al. Performance of an abbreviated version of the Lubben Social Network Scale among three European community-dwelling older adult populations. Gerontologist [Internet]. 2006;46(4):503-513. doi: https://doi.org/10.1093/geront/46.4.503

10. Kurimoto A, Awata S, Ohkubo T, Tsubota-Utsugi M, Asayama K, Takahashi K, et al. Reliability and validity of the Japanese version of the abbreviated Lubben Social Network Scale. Nippon Ronen Igakkai Zasshi [Internet]. 2011;48(2):149-157. doi: https://doi.org/10.3143/geriatrics.48.149

11. Ribeiro O, Teixera L, Duarte N, Azevedo MJ, Araújo L, Barbosa S, et al. Versão portuguesa da escala breve de redes sociais de Lubben (LSNS-6). Revista Kairós-Gerontología [Internet]. 2012;15(11):217-234. Recuperado a partir de: https://revistas.pucsp.br/index.php/kairos/article/view/12787

12. Lee KW, Kim SY, Chung W, Hwang GS, Hawang YW, Hwang IH. The Validity and Reliability of Korean Version of Lubben Social Network Scale. Korean J Fam Med [Internet]. 2009;30(5):352-358. doi: http://doi.org/10.4082/kjfm.2009.30.5.352

13. Myagmarjav S, Burnette D, Goeddeke FJ. Comparison of the 18-item and 6-item Lubben Social Network Scales with community dwelling older adults in Mongolia. PLoS One [Internet]. 2019;14(4):e0215523. doi: https://doi.org/10.1371/journal.pone.0215523

14. Chang Q, Sha F, Chan CH, Yip PSF. Validation of an abbreviated version of the Lubben Social Network Scale (“LSNS-6”) and its associations with suicidality among older adults in China. PLoS One [Internet]. 2018;13(8):e0201612. doi: https://doi.org/10.1371/journal.pone.0201612

15. Buckley T, Becker T, Burnette D. Validation of the abbreviated Lubben Social Network Scale (LSNS-6) and its association with self-rated health amongst older adults in Puerto Rico. Health Soc Care Community [Internet]. 2022;30(6):e5527-38. doi: https://doi.org/10.1111/hsc.13977

16. Nunnally JC. Psychometric Theory (2nd ed.). New York: McGraw-Hill; 1978.

17. Moyano-Díaz E, Ramos-Alvarado N. Bienestar subjetivo: midiendo satisfacción vital, felicidad y salud en población chilena de la Región Maule. Universum (Talca) [Internet]. 2007;22(2):177-193. doi: http://dx.doi.org/10.4067/S0718-23762007000200012

18. Moyano-Díaz E, Dinamarca D, Mendoza-Llanos R, Palomo-Vélez G. Escala de Felicidad para Adultos (EFPA). Ter Psicol [Internet]. 2018;36(1):37-49. doi: http://dx.doi.org/10.4067/s0718-48082017000300033

19. Andersson BT, Christensson L, Fridlund B, Broström A. Development and psychometric evaluation of the radiographers’ competence scale. Open Journal of Nursing [Internet]. 2012;2:85-96. doi: http://dx.doi.org/10.4236/ojn.2012.22014

20. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Modeling [Internet]. 1999;6(1):1-55. doi: https://doi.org/10.1080/10705519909540118

21. Bandalos DL, Finney SJ. Factor analysis: Exploratory and confirmatory. En: Hancock GR, Mueller RO, editores. The reviewer’s guide to quantitative methods in the social sciences [Internet]. New York: Routledge; 2010. p.93-114. doi: https://psycnet.apa.org/doi/10.4324/9781315755649-8

22. Cheung GW, Rensvold RB. Evaluating goodness-of-fit indexes for testing measurement invariance. Struct Equ Modeling [Internet]. 2002;9(2):233-255. doi: https://doi.org/10.1207/S15328007SEM0902_5

23. Kenny DA, Kaniskan B, McCoach B. The performance of RMSEA in models with small degrees of freedom. Sociol Methods Res [Internet]. 2014;44(3):486-507. doi: https://doi.org/10.1177/0049124114543236

24. Encuesta Bicentenario. Resultados Encuesta Nacional Bicentenario 2023 [Internet]. Santiago: Pontificia Universidad Católica de Chile; 2023. Available from: https://encuestabicentenario.uc.cl/publicaciones/resultados-encuesta-bicentenario-2023/

25. Dimitrov DM. Testing for Factorial Invariance in the Context of Construct Validation. Meas Eval Couns Dev [Internet]. 2010;43(2):121-49. doi: https://doi.org/10.1177/0748175610373459

26. Bentler, P. M. EQS 6 Structural Equations Program Manual. Encino, CA: Multivariate Software Inc; 2006.

Notas de autor

eemoyano@gmail.com

Información adicional

How to reference. : Moyano-Diaz E, Jorquera-Cox ME, Oda-Montecinos CM, Estrada-Goic C. Validation of the Lubben-6 Scale to measure social support network in older adults in Chile. MedUNAB [Internet]. 2025;28(2):345-353. doi: https://doi.org/10.29375/01237047.5070

Autor Contributions: EMD. Conceptualization: formulation and design of the study and contribution to obtain funding. Methodology: development of the research protocol and methodological design. Formal analysis: data analysis and synthesis. Investigation: guidance on sample selection procedures and data collection. Writing: drafting of original, final, and revised manuscript. Supervision: project leadership. Funding acquisition: obtaining financial resources. MEJC, CMOM, and CEG. Investigation: sample selection and survey administration. Revision of manuscript drafts. Resources: data collection and submission from each city.

Información adicional

redalyc-journal-id: 719

Enlace alternativo

https://revistas.unab.edu.co/index.php/medunab/article/view/5070 (html)

https://revistas.unab.edu.co/index.php/medunab/article/view/5070/4287 (pdf)

https://revistas.unab.edu.co/index.php/medunab/article/view/5070/4288 (pdf)

https://revistas.unab.edu.co/index.php/medunab/article/view/5070/4316 (xml)