Study Design
This cross-sectional study employed a nonprobabilistic sampling approach. The data reported in this article were derived from the Longitudinal Older Learners (LOL-Costa Rica) project database, which was developed through international collaboration between the School of Psychology at the Universidad Nacional in Costa Rica and the Universidad de Valencia in Spain.
Participants
The inclusion criteria for participation in the study were age ≥ 50 years and enrollment in a program for older adults at the Universidad Nacional of Costa Rica (PAIPAM-UNA). This programme offers courses that provide training on aging-related topics for comprehensive development and well-being. The target population of this program is concentrated in the province of Heredia (metropolitan-urban area).
A total of 282 Costa Ricans aged ≥ 50 years (M=64, SD=7.55) participated in the study, of whom 91% were women. This high female participation aligns with various studies conducted in the country among older adults in university programs [25, 26, 27, 28]. Regarding educational level, 38.1% had university education, 32.4% had secondary education, and 29.6% had primary education. The marital status of the participants indicated that 46.8% of the older adults were married, 21.3% were divorced, 21% were widowed, and 10.3% were single. A total of 88.9% of the participants lived in their homes.
Both the instruments and informed consent forms were administered in paper format. The participants voluntarily completed the instruments, and the data were treated anonymously.
Instruments
SF-8 Health-Related Quality of Life Scale
The SF-8 scale is a shortened version of the SF-36 scale developed by Ware et al. [29]. It constitutes a generic assessment of the cognitive and physical aspects of health-related quality of life [15]. Despite having similar characteristics and behaviors, this version is less widely used and concise. The instrument is typically structured into two factors, one physical and one cognitive, for which reliabilities of .85 and .79, respectively, have been reported [13]. However, some studies have indicated a better fit with these three factors (adding a general health factor) [19, 15]. Its items are of the Likert type, ranging from 1 to 5, with reverse scoring; that is, a lower score indicates a better perception of health.
Temporal Satisfaction with Present Life Scale
The Temporal Satisfaction with Life Scale (TSLS) [30] measures overall life satisfaction regarding the past, present, and future aspects of an individual's life. For this study, only five items related to the subscale of satisfaction with present life were used. Its adaptation to Spanish demonstrated appropriate psychometric properties, including evidence of construct and convergent validity. It consists of Likert-type items ranging from 1 (strongly disagree) to 5 (strongly agree) [31]. The reliability of the 15-item version of the instrument typically falls between .87 and .93, according to Cronbach´s alpha [30, 32], and between .81 and .86 in its Spanish version [33]. The instrument provided convergent validity evidence for the SF-8 [31].
Duke-UNC-11 Functional Social Support Questionnaire
The original version of this instrument was found in The Duke-UNC Functional Social Support Questionnaire [34]. However, in this study, a Spanish adaptation was used [35]. Duke-UNK-11 assesses perceived functional social support. The instrument comprises two dimensions: confidential support (having others to communicate with) and affective support (having demonstrations of affection, love, and empathy). It consists of 11 Likert-type items, ranging from 1 to 5. This instrument was used to provide evidence of convergent validity because it not only serves as a subjective indicator of successful aging but also because social disconnection is a relevant aspect linked to health perception [36]. There is evidence of significant correlation between SF-8 and Duke-UNC-11 [37].
Revised Lubben Social Network Scale (LSNS-R)
The revised version developed by Lubben et al. [38] measures the size, closeness, and frequency of social contact among older adults. In this study, a Spanish adaptation was used [39].
Additionally, there is convergent validity evidence between the SF-8 and social support for both the family and friends subscales [37]. The LSNS-R comprises 12 Likert-type items that distinguish between the social networks of friends and family. Participants can score between 0 and 60 points in total or between 0 and 30 on each subscale. A higher score indicates a higher level of support from the social network [40, 41].
Collection Procedures
Students from PAIPAM-UNA courses were invited to participate. The researchers explained the project, the purpose of the research, how the data would be gathered, and the voluntary, anonymous, and confidential nature of participation, emphasizing that the data would be used solely for scientific purposes.
Individual interviews were planned with older adults who were unable to respond to the questionnaire independently. Finally, the surveys were administered along with informed consent. Sociodemographic variables were assessed, along with the rest of the psychosocial items and instruments.
Statistical Analysis
Confirmatory factor analysis was conducted to assess the fit of the theoretical structure of the SF-8 health-related quality of life scale in the utilized sample, initially testing the traditional 2-factor structure and subsequently the 3-factor structure. A comparative fit index (CFI) above .90, root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR) for observed errors below .08 were considered indicators of good fit [42].
Internal consistency was examined using Cronbach's alpha for the SF-8 factors and instruments, aimed at providing evidence of nomological validity. Bivariate Pearson correlations were then performed to evaluate convergent validity. The data composition did not allow for invariance analysis between gender and other sociodemographic variables because of limitations in the number of parameters in the Confirmatory Factor Analysis theoretical model. For instance, in the case of gender, the sample size for men (N=19) was insufficient compared to the required number of parameters, preventing the analysis. Two software programs were employed for the analyses: IBM SPSS version 25 (IBM Corp., 2017) and the statistical software Jamovi, v. 2.3.18.0 (Jamovi Project, 2023).