Introduction
The public image of older adults is undergoing a significant transformation – the stereotype of passive, withdrawn seniors is increasingly being replaced by a perception of active, engaged and life-loving ‘silver’ individuals (Formosa Reference Formosa2019; Phillipson et al. Reference Phillipson, Bernard and Strang2024).
This shift is driven by both objective and subjective factors. Objective ones include improved living and working conditions, medical and technological advances, and longer life expectancy (Reid Reference Reid2016). Subjective factors involve greater wellbeing, emerging aspirations, the need for self-actualization, and sustained social relationships (Ryff Reference Ryff and Waterman2013; Yazdanpanahi et al. Reference Yazdanpanahi, Pantelaki, Holland, Gilroy, Spencer, Weston and Rogers2024).
In this context, Universities of the Third Age (U3As), which have been operating successfully in many European countries for over five decades (Formosa Reference Formosa2019), have gained particular significance. These U3As not only promote the active engagement of older adults but, more importantly, address their diverse needs – ranging from educational and cognitive to social, emotional and developmental. As a result, they contribute to enhanced quality of life and the maintenance of wellbeing in later adulthood (Swindell Reference Swindell1993; Villar et al. Reference Villar, Pinazo, Triadó, Celdrán and Solé2010; Gierszewski and Kluzowicz Reference Gierszewski and Kluzowicz2021).
In Poland, U3As originated in 1975, but their rapid development occurred during the first two decades of the 21st century. Their activities are modelled on British and French solutions, adapted to the country’s cultural context. The aim of U3As is to support older adults through educational activities such as expanding knowledge and stimulating intellectual engagement; and psychotherapeutic and integrative functions such as enhancing wellbeing, reducing isolation and facilitating social contact (Fabiś and Konieczna-Woźniak Reference Fabiś and Konieczna-Woźniak2019).
Polish U3A students are mostly baby boomers who experienced post-war hardships and decades under a socialist regime, including martial law. This generation contributed significantly to the political transformation of 1989, is aware of its agency, yet remains strongly attached to traditional values. It also represents the most religious segment of Polish society, with a strong work ethic and a deep connection to the land (Wątroba Reference Wątroba2022).
The specificity of older adults’ needs
Needs may be defined as conditions or states that enhance an individual’s wellbeing (Dean Reference Dean2010). They include both basic biological necessities that are essential for survival, such as food and oxygen, and subjective factors shaped by experience and learning, such as the need for achievement, recognition or social connection. The term ‘need’ can also refer to a state of deprivation that motivates action; the absence of food or social bonds can drive behaviour aimed at fulfilling these needs.
A valuable framework for analysing older adults’ functioning is Alderfer’s (Reference Alderfer1972) ERG theory, which classifies needs into three categories: existence, relatedness and growth. Existence needs involve material and physiological factors, such as safe housing and living conditions. Relatedness refers to forming and maintaining satisfying relationships with family and friends. Growth encompasses self-actualization, including the development of talents, creativity and competence through goal attainment.
Unlike Maslow’s hierarchy, ERG theory allows greater flexibility – needs from different levels can be fulfilled simultaneously. It also assumes the possibility of regression when higher-level needs are frustrated, and that higher needs may motivate even if lower needs are only partially met. These principles align with observations about older adults.
As noted by gerontologists and developmental psychologists (Stuart-Hamilton Reference Stuart-Hamilton2012; Bruggencate et al. Reference Bruggencate, Luijkx and Sturm2018), psychosocial needs gain increasing importance with age. Motivators such as the need for social acceptance, belonging and emotional bonds that provide meaning begin to dominate the hierarchy. Equally significant are needs for security, respect and a sense of usefulness, often expressed through helping loved ones – for example caring for grandchildren – and passing on life wisdom to younger generations, reflecting a concern for supporting the younger generation (Erikson Reference Erikson1980).
McClusky (1963, cited in Merriam Reference Merriam1987) proposed an alternative classification of older adults’ needs, identifying the following categories: (1) coping needs, which pertain to managing changes and challenges associated with ageing, such as health issues, the loss of loved ones and adjustment to new life stages; (2) expressive needs, which involve engaging in activities that allow individuals to express themselves and pursue creative endeavours that yield personal fulfilment; (3) contributive needs, reflecting the desire to contribute to society and support others; (4) influence needs, which relate to the wish to feel influential and participate in decisions that affect one’s life and the broader community; and (5) transcendence needs, encompassing the aspiration to move beyond the self, seek meaning and purpose, and connect with a higher power or the Absolute.
Furthermore, according to self-determination theory (Ryan and Deci Reference Ryan and Deci2000, Reference Ryan and Deci2017), individuals have an innate capacity to choose developmental activities that satisfy core psychological needs. These include autonomy – the ability to act in line with personal values; competence – the capacity to achieve goals and influence one’s environment; and relatedness – a sense of social belonging and acceptance. Human functioning is largely shaped by context, and the social environment plays a key role in supporting these needs. Thus, U3As focus on creating conditions that fulfil these universal psychological needs, thereby enhancing the wellbeing of older adults.
Older adults’ needs fulfilled through U3A classes
Non-formal educational institutions such as U3As offer older adults opportunities to meet a wide range of needs in a meaningful and fulfilling manner. Retirement, along with changes in the socio-cultural environment, creates favourable conditions for continued learning and personal development among older adults. These circumstances motivate older adults to explore new areas of interest and adapt to an ever-changing reality (Hoare Reference Hoare2011; Istance Reference Istance2015).
Late adulthood is a period of reflection on one’s life, reviewing achievements and making decisions that are important for further development. Older adults’ activities go beyond the intellectual domain – they include not only the pursuit of knowledge and skills but also the need to build peer relationships (Bruggencate et al. Reference Bruggencate, Luijkx and Sturm2018; Noble et al. Reference Noble, Medin, Quail, Young and Carter2021). Importantly, older adults seek recognition and social approval for their actions. Among the key needs met through participation in U3A classes are social and cognitive needs (Findlay Reference Findlay2003).
The aspiration for independence, self-sufficiency and active engagement in social life is a key factor in effective adaptation and maintaining a sense of control over one’s life and environment – both social and physical (Hoare Reference Hoare2011; Dostalova et al. Reference Dostalova, Bártová, Blahova and Holmerova2021). The U3As support this process by offering activities that promote not only knowledge and skills development but also social integration, helping older adults’ adapt to the new roles and challenges of late adulthood (Formosa Reference Formosa2019).
As part of their activities, U3As also organize educational trips, excursions and cultural events. These initiatives broaden participants’ intellectual horizons and encourage the exploration of new areas of knowledge. Participation in such activities brings satisfaction to older adults by addressing their cognitive and personal development needs (Formosa Reference Formosa2014; Jenkins and Mostafa Reference Jenkins and Mostafa2015).
Physical activity is a permanent component of the educational programmes offered by most U3As. Medical experts (Herbert et al. Reference Herbert, Meixner, Wiebking and Gilg2020) emphasize that regular exercise plays a vital role in maintaining both mental and physical wellbeing, regardless of age. Health care, therefore, ranks among the most important needs addressed through U3A participation (Formosa Reference Formosa2014, Reference Formosa2019).
Older adults themselves often serve as initiators and organizers of such activities, particularly through their involvement in U3A student councils. This engagement enables them to influence the range of offerings and fulfil their need for agency. Volunteer initiatives coordinated by U3As further support both a sense of influence and personal development (Cocca-Bates and Neal-Boylan Reference Cocca-Bates and Neal-Boylan2011).
A systematic review and meta-analysis by Noble et al. (Reference Noble, Medin, Quail, Young and Carter2021) confirms that older adults’ participation in both formal and informal education positively affects cognitive functioning and wellbeing. Based on 32 studies, the authors estimated an average difference of 0.40 points on the Mini-Mental State Examination between participants and control groups. Educational engagement was also linked to improvements in older adults’ sense of agency, belonging and overall wellbeing.
Practical value of an inventory for assessing need fulfillment among older adults
Meeting human needs is fundamental to proper functioning and a fulfilling life, regardless of age, gender, origin or social status (Ryan and Deci Reference Ryan and Deci2000). In the case of older adults, various institutions – such as U3As – play an important role in this area, bringing together an active, engaged and often more emancipated segment of the oldest generation. The specific characteristics of this group require appropriately tailored educational activities that reflect their actual needs, expectations and capabilities.
In particular, the dynamic growth in the number of U3A institutions in Poland and worldwide highlights the importance of professionalizing the process of diagnosing the diverse needs of older adults and the extent to which they are satisfactorily met. In this context, it is essential to use a standardized diagnostic tool with proven reliability and validity, enabling their credible and objective assessment.
The early implementation of such a tool within an institution allows the educational offer to be tailored to the actual needs of its recipients and to their expectations regarding the extent to which these needs are satisfactorily met. Programmes developed on this basis can both meet participants’ expectations – providing a sense of meaning and satisfaction from intellectual engagement – and prevent disappointment, frustration or a loss of trust in educational institutions.
The dynamic growth of U3As also presents opportunities for ongoing programme adjustments, experimentation with new teaching methods, the creation of inclusive learning environments and adaptation of formats to evolving participant expectations. However, the effectiveness of these efforts depends on a reliable diagnosis of need fulfilment among beneficiaries, supported by a validated and appropriate measurement tool.
A well-tailored educational offer can significantly improve the quality of life of older adults by strengthening their sense of belonging, enhancing family relationships, fostering involvement in the local community and supporting personal development. Understanding specific educational needs also enables expanding the offer and reaching a broader audience, including those previously undecided or unconvinced by earlier U3A proposals. Moreover, using an accurate diagnostic tool creates real opportunities to secure external funding. Reliable assessment of the level of specific need fulfilment improves the preparation of grant applications, increasing the likelihood of positive evaluation and funding.
So far, no comprehensive multidimensional inventory of needs fulfilment among U3A students has been developed. Given the importance of understanding the extent to which selected needs are satisfied for planning an effective U3A educational offer, a pioneering effort has been made to construct and validate a research tool for this purpose. Though designed with Polish audiences in mind, this tool can be adapted for use in other countries and cultures.
Aim of the studies
The main aim of Studies 1 and 2 was to develop a multidimensional inventory of needs fulfilled by older adults during U3A activities. The instrument was constructed in several stages following an inductive–deductive strategy (Creamer Reference Creamer2017; Hall Reference Hall2020).
In Study 1, based on qualitative material from interviews with U3A participants, statements for a preliminary version – the Need Fulfillment Inventory for U3A Students (NFI-72) – were generated. Items with unsatisfactory psychometric properties were subsequently removed. As a result, a shortened final version (NFI-30) was obtained, whose structure was verified for factor validity, reliability (internal consistency) and item discriminative power.
In Study 2, the theoretical validity of the instrument (convergent and discriminant) and its reliability defined as temporal stability was assessed.
Research procedure for Study 1
In Study 1, a preliminary 72-item version of the multidimensional NFI-72 was developed. Subsequently, using relevant statistical procedures, the final 30-item version of the instrument (NFI-30) was proposed.
Method
Participants
The sample was recruited using voluntary convenience sampling (Privitera Reference Privitera2022). Participants were eligible if: (1) U3A authorities approved the study during U3A-organized classes, (2) they gave informed consent, and (3) they attended classes on the study day.
A total of 1,097 older adults completed the questionnaire (85% females, 13% males, 2% no data), all affiliated with 22 U3As in Silesia (19) and Małopolska (3) voivodeships in Poland. Minimum sample sizes required to assess factor structure of the NFI-72 and the NFI-30 were calculated using G*Power v. 3.1.9.7 software. Assuming f2 = 0.15, α err prob = 0.05, power = 0.95 and number of predictors = 72 or 30, Nmin amounted to 381 and 260, respectively.
Data utilized in exploratory factor analysis to evaluate the NFI-72 were obtained from a sample of 660 U3A students, substantially exceeding Nmin (ratio = 1.73). In turn, data employed in confirmatory factor analysis to evaluate the NFI-30 were derived from 437 U3A participants, also substantially exceeding Nmin (ratio = 1.68).
The mean age was 71.362 years (SD = 5.475). Regarding residence: 9 per cent lived in rural areas, 13 per cent in towns of 10,000–30,000 residents, 18 per cent in 31,000–50,000, 7 per cent in 51,000–100,000 and 45 per cent in cities with over 100,000 residents (8 per cent no data). Participants reported vocational (38%), secondary (42%) or higher (13%) education, with 8 per cent with no data. Most (83%) were retired, on average for 11.853 years (SD = 7.340). In terms of marital status, 31 per cent were widowed; the rest were in a relationship (32%) or single (26%); no data (10%).
According to official statistics (GUS 2025), 86,600 individuals in Poland participate in activities offered by U3As, with 552 such organizations in operation. In terms of distribution, the Silesia and Małopolska voivodeships rank second and sixth nationally, respectively, with nearly 100 U3As located there combined.
Females constitute 86 per cent of participants, most of whom are retirees, typically in their 70s and 80s. Half have completed secondary education, while over one-third hold tertiary qualifications. Participation requires sufficient financial resources to cover membership fees and travel costs, as well as adequate health for regular attendance. In this respect, U3A participants represent a relatively selective segment of the older population; within Poland’s population of 37 million, around one in every 100 individuals aged 60 and over is involved in U3A activities.
Given these characteristics, the study sample, although non-random, appears to reflect key attributes of the U3A population, including gender, age, education and place of residence. With a sample size (n = 1,097), it represents 1.3 per cent of the target population. Data were collected across 22 U3As, accounting for approximately 4 per cent of such organizations in Poland.
The NFI design
The tool was developed in four stages, using an inductive–deductive strategy (Creamer Reference Creamer2017). In the first stage, three judges (a psychologist specializing in ageing and two pedagogues engaged in research on adult education and geragogy) independently analysed transcripts of qualitative interviews previously conducted with 16 U3A students (13 females and 3 males). Participant selection was non-random and voluntary. The sample included U3A students from centres in the Silesian and Małopolskie voivodeships who met the following criteria: aged over 60; at least five years of active participation in U3A classes; and sufficient communicative competence.
The interviews aimed to explore needs related to development in later adulthood in the context of U3A participation, focusing on participants’ meanings of later-life development, perceived needs, ways of fulfilling them and expectations regarding engagement in U3A activities. Interviews lasted 50–80 minutes. The opening question was: ‘Please tell me about your life from the perspective of your educational experiences’, followed by: ‘What does U3A mean to you?’ and ‘What does participation in U3A give you?’. Ethical standards were adhered to throughout the study.
The judges were instructed to: (1) identify, within the qualitative material, statements by older adults concerning the needs fulfilled in U3A classes; (2) generate as many paraphrases of these excerpts as possible; (3) organize the excerpts and their corresponding paraphrases; (4) derive ad hoc need categories (e.g. social influence, knowledge acquisition, recreation) based on the excerpts and paraphrases; and (5) assign the excerpts and paraphrases to the identified categories.
In total, the judges identified 164 excerpts and 257 corresponding paraphrases (on average, 55 excerpts and 86 paraphrases per judge), which were subsequently classified into six preliminary need categories: (1) social, (2) cognitive, (3) health-related, (4) adaptive, (5) control and (6) self-development.
In the second stage, a different group of three independent judges (two psychologists specializing in ageing and a pedagogue specializing in geragogy) used the merged excerpts of older adults’ statements and paraphrases to generate items describing the six previously identified categories of needs fulfilled in U3A classes. Following instructions, each item began with: ‘Classes at U3A …’, for example ‘Classes at U3A … prevent me from isolating myself at home’. The judges generated 538 preliminary statements. Items were generated to achieve theoretical saturation of the six previously identified categories of U3A students’ needs – that is, the point at which the content of additional items no longer contributed new, meaningful information to the need categories.
Next, the three judges from the first stage independently evaluated these 538 statements. Their assessments were based on two criteria: (1) which of the six identified needs the statement reflected and (2) how representative it was of that need. The first was rated on a nominal scale: social – 1, cognitive – 2, health – 3, adaptation – 4, control – 5, self-development – 6. The second was rated on an ordinal scale from 1 (poorly representative) to 6 (highly representative)
The judges used the following working definitions of needs when evaluating the statements: (1) social needs involve the desire to be with others, build satisfying relationships, make friends and engage socially; (2) cognitive needs refer to participation in mental activities, seeking knowledge, analysing situations, solving problems and learning; (3) self-development needs involve efforts to improve oneself and enhance personal resources; (4) health needs relate to caring for physical and mental wellbeing to maintain functional fitness; (5) adaptation needs reflect acquiring skills and competencies to meet changing environmental demands and stay current; (6) control needs involve self-efficacy, belief in achieving goals, initiative and influence over one’s social and physical environment (Heckhausen Reference Heckhausen2000; Demick and Andreoletti Reference Demick and Andreoletti2003; Dean Reference Dean2010; Hoare Reference Hoare2011; Stuart-Hamilton Reference Stuart-Hamilton2012; Verhaeghen and Hertzog Reference Verhaeghen and Hertzog2014; Bruggencate et al. Reference Bruggencate, Luijkx and Sturm2018).
In the third stage, the judges analysed statements best representing a specific need (criterion 1 for statement evaluation). The cut-off point was set at a mean score of 5.333 (75th percentile or higher), based on all ratings. Among the 150 selected statements, 47, 25, 19, 7, 17 and 20 represented social, cognitive, health, adaptation, control and self-development needs, respectively.
Next, the degree to which each of the 150 statements was consistently classified by judges into one of the six need categories (criterion 1) was assessed. Classification consistency was high, with Krippendorff-α = 0.916 (PU 95%: 0.872–0.953) (Krippendorff Reference Krippendorff2022).
Statements with repetitive content were excluded, leaving only the one judged most representative of each need. For example, from: Classes at U3A … ‘allow me to get involved in helping others’, ‘allow me to act for others’ and ‘are the place where I can help others’, the last was included in the NFI. Redundant items were excluded, retaining these statements with unique content, through an open discussion among the judges. In this way, the number of statements for each need was reduced to 12, except for adaptation, where five missing items were created based on need theory and relevant research (Horvath Reference Horvath1998; Dean Reference Dean2010; Ring et al. Reference Ring, Villar, Ristolainen and Tiilikainen2025), for example Classes at U3A … ‘help me feel familiar with new advancements, such as technological ones’.
Ultimately, the preliminary version of the inventory (NFI-72) contained 72 statements evenly distributed across six need categories (see Table 1). Items were rated on a five-point scale: 1 – definitely not, 2 – rather not, 3 – no opinion, 4 – rather yes, 5 – definitely yes. A section on basic socio-demographic characteristics of older adults was added to the main part of the NFI-72.
The EFA results for the NFI-72 and the NFI-30
Table 1 Long description
Exploratory factor analysis results are listed for two inventories, NFI-72 and the shorter NFI-30, showing each item’s assigned domain, four factor loadings, and item uniqueness. In NFI-30, development items load most strongly on Factor 1, with the highest loading for item i52 (i1) at about 0.75, and other development items mostly around 0.45 to 0.68. Social items in NFI-30 cluster on Factor 2, with loadings roughly 0.60 to 0.68, while agency items cluster on Factor 3 with the strongest loading about 0.79 for item i10 (i3). Health items in NFI-30 cluster on Factor 4, with several high loadings around 0.72 to 0.78, and one health item (i63 (i24)) showing a more moderate loading near 0.41. In NFI-72, many control and adaptation items load strongly on Factor 1, including a top loading about 0.81 for “help me solve my problems,” while education and self-development items tend to load more on Factor 2 and social-connection items more on Factor 3; physical-activity items load highest on Factor 4, with a peak around 0.81. Eigenvalues indicate a dominant first factor in NFI-72 (about 25) compared with a smaller first factor in NFI-30 (about 11), while variance explained is more evenly distributed across the four factors in NFI-30 (about 11 to 13 percent each). Some items show notable cross-loadings or negative secondary loadings, so factor labels should be interpreted as dominant tendencies rather than exclusive categories.
Notes: ADA – adaptation, AGE – agency, COG – cognitive, CON – control, DEV – development, HEA – health, SDEV – self-development, SOC – social.
In the ‘Item (indicator)’ column for the NFI-30, numbers from the final version of the inventory are given in parentheses.
Factor loadings for the items included in the NFI-30 are presented in bold.
Research procedure
First, permission to conduct the survey was obtained from U3A authorities. After approval, the paper version of the tool was distributed to older adults during selected classes. Participants were informed about the study’s purpose, assured of voluntary participation and reminded of their right to withdraw at any time. Before the survey began, it was confirmed that the questionnaire items and instructions were clear and understandable.
Students completed the questionnaire during classes, immediately after the paper version had been distributed. As a result of this procedure, a 100 per cent response rate was achieved – all individuals who agreed to participate returned a completed questionnaire, although, approximately 3–5 per cent of the data were missing. The procedure was approved by the Research Ethics Committee of the authors’ home university (opinion no. AWSB/KE/1/2025).
Results
The NFI-72: factor structure evaluation – EFA results
To assess factor validity of the working version of the tool (72 items), it was assumed that the statements reflect six need categories: social, cognitive, health, adaptation, control and self-development. The assumed structure of need fulfilment was tested using exploratory factor analysis (EFA), based on data from 660 U3A students. Due to the non-normal (left-skewed) distributions of the variables and the highly likely correlations among items representing the identified need categories (e.g. adaptation and control, cognitive and self-development), the principal axis factoring method with PROMAX rotation was applied (Fabrigar and Wegener Reference Fabrigar and Wegener2012).
Preliminary tests – the Kaiser–Meyer–Olkin (KMO) measure and Bartlett’s test – indicated that correlations among variables were sufficiently strong to justify dimensionality reduction in the conducted analysis (KMO = 0.965; χ2(2556) = 26,909.995; p < 0.001, respectively; Elliott and Woodward Reference Elliott and Woodward2014).
Contrary to expectations, the EFA revealed a four-factor structure (see Table 1). Factor I explained 16 per cent of variance and grouped items related to adaptation and control needs. Factor II explained 13 per cent and correlated with cognitive and self-development items. The final two factors explained 9 per cent and 6 per cent of variance, corresponding to social and health needs, respectively.
Given the results indicating strong content similarity between control and adaptation needs (e.g. items 10 and 35), as well as between self-development and cognitive needs (e.g. items 52 and 57), these categories were merged into two broader ones: agency and development needs, respectively. This reduced the number of need categories to four: (1) agency (AGE factor), (2) development (DEV factor), (3) social (SOC factor) and (4) health (HEA factor).
The four-factor structure demonstrated satisfactory model fit (RMSEA = 0.053; 90% CI: 0.051–0.054; SRMR = 0.036) and conceptual clarity, as the instrument items loaded clearly on the intended factors without substantial contamination from other factors. Alternative structures, namely the initial six-, five- and three-factor solutions, were also tested. In each case, model fit was comparable to the four-factor structure; however, their clarity and interpretability were substantially reduced.
For the six-factor solution (RMSEA = 0.049; SRMR = 0.030), the eigenvalue of Factor VI was 0.911. Assuming a factor loading of ≥ 0.40 as the cut-off criterion (Stevens Reference Stevens2002; Hair et al. Reference Hair, Black, Babin and Anderson2009), Factor I comprised items representing control, adaptation, health, social and self-development needs (10, 6, 3, 2 and 2 items, respectively). Factor II correlated with items theoretically related to cognitive and self-development needs (9 and 6 items, respectively). Factor III was primarily associated with social needs (9 items), whereas Factor IV mainly reflected health needs (6 items). Factors V and VI together were associated with four items reflecting adaptation, control and self-development needs.
In the five-factor structure (RMSEA = 0.050; SRMR = 0.033; eigenvalues ranging from 1.049 to 25.110), many items still exhibited cross-loadings. Factor I was defined by items related to control, adaptation, health, self-development and social needs (11, 9, 4, 3 and 1 items, respectively). Factor II consisted primarily of items reflecting cognitive needs (9 items), followed by self-development, social and adaptation needs (5, 3 and 1 items, respectively). Factor III correlated with 10 social-needs items and 1 health-needs item, whereas Factor IV was associated exclusively with health-related content (6 items). Factor V contained only two items, both reflecting self-development needs.
Finally, in the three-factor structure (RMSEA = 0.058; SRMR = 0.042; eigenvalues ranging from 1.920 to 25.071), Factor I correlated with items reflecting cognitive and self-development needs (9 items each), followed by health (6 items) and social, adaptation and control needs (2, 1 and 1 items, respectively). Factor II comprised items related to control and adaptation needs (11 and 9 items, respectively), followed by self-development, health and social needs (3, 2 and 1 items, respectively). Factor III was associated with social-needs items (9 items) as well as adaptation, control and health needs (1 item each).
The NFI-30: factor structure evaluation – EFA results
For further analysis, items most strongly correlated with need types within each factor were selected. Considering the sample size (n > 300) and the questionnaire-based nature of the study, a factor loading ≥ 0.40 was adopted as the cut-off (Stevens Reference Stevens2002; Hair et al. Reference Hair, Black, Babin and Anderson2009). Items with lower or cross-loadings were deemed unrepresentative of the four identified need categories and excluded from further analyses. This resulted in a more coherent and stable factor structure, with clearer factor separation and greater interpretive clarity (Stevens Reference Stevens2002; Fabrigar and Wegener Reference Fabrigar and Wegener2012).
Ultimately, eight statements represented the AGE factor (four reflecting control and four related to adaptation needs), eight the DEV factor (four related to cognitive and for reflecting self-development needs), eight the SOC factor, and six the HEA factor – 30 items in total. An additional exploratory factor analysis (principal axis method, PROMAX rotation; KMO = 0.946; Bartlett’s χ2(435) = 9608.480; p < 0.001) conducted on these 30 items confirmed the assumed four-factor structure of need fulfilment among U3A students (see Table 1).
All items showed distinct correlations with their respective factors. The DEV, SOC, AGE and HEA factors explained 13 per cent, 13 per cent, 12 per cent and 11 per cent of the variance, respectively. Further analyses were thus conducted on the 30-item version of the inventory (NFI-30), assuming a four-factor model.
The NFI-30: factor validity – CFA results
To verify the factor validity of the instrument, confirmatory factor analysis (CFA) was performed (see Table 2) using data from 437 U3A students who were not previously included in the EFA sample (n = 660). Due to the violation of the multivariate normality assumption (Mardia’s index = 219.9), maximum likelihood estimation with robust standard errors (MLR) was applied (Hoyle Reference Hoyle2023).
Reliability, item discrimination power of the NFI-30, and the CFA results
Table 2 Long description
Reliability and confirmatory factor analysis results are reported for four latent factors (DEV, SOC, AGE, HEA) and a global score, listing each item’s alpha if dropped, item to rest correlation, and CFA loading estimates with confidence intervals. Global internal consistency is 0.943, while factor reliabilities are 0.844 for DEV, 0.817 for SOC, 0.853 for AGE, and 0.839 for HEA. Item to rest correlations range from 0.274 (SOC item i1) up to 0.669 (AGE item i35), with an average of 0.582, indicating generally moderate discrimination with one notably weak SOC item. Standardized CFA loadings are all positive and statistically significant, spanning roughly 0.531 to 0.818; the strongest loadings appear in HEA (about 0.680 to 0.818) and DEV (about 0.610 to 0.777), while the lowest loading is AGE item i25 at about 0.531. Confidence intervals for standardized loadings are fairly tight overall, suggesting stable estimates across items. Intercorrelations among factors are moderate to strong (about 0.549 to 0.759), and each factor correlates strongly with the global score (about 0.792 to 0.865), supporting a coherent overall construct. Because the table summarizes psychometric fit and associations, it supports internal consistency and factor structure but does not by itself establish causality or external validity.
Notes: In the ‘Item (indicator)’ column, numbers from the final version of the inventory are given in parentheses.
Cronbach’s α values for the parts of the inventory (factors) and for the overall scale (global) are presented in bold on the diagonal of the matrix.
Correlation coefficients between factors are shown above the diagonal.
Covariance values between factors are shown below the diagonal.
The item content is presented in Table 1.
Significance level: ** p < 0.01.
The proposed NFI structure was confirmed, with the tested model showing satisfactory fit: χ2(383) = 812.400; p < 0.01; CFI = 0.912; TLI = 0.900; NFI = 0.848; RMSEA = 0.051, 90% CI: 0.047–0.055; SRMR = 0.058; GFI = 0.968 (Tinsley and Brown Reference Tinsley and Brown2000).
As expected, all items were significantly and positively associated with their respective factors, and the strength of these relationships was moderate to high (Tinsley and Brown Reference Tinsley and Brown2000). Standardized loadings (λ) for items reflecting development needs ranged from 0.610 to 0.777. Comparable values were found for social (0.553–0.738), agency (0.531–0.748) and health needs (0.680–0.818). Furthermore, all dimensions were positively intercorrelated, with correlations ranging from r = 0.549 (SOC–HEA) to r = 0.759 (DEV–SOC).
The NFI-30: evaluation of reliability (internal consistency)
The shortened version of the tool demonstrated high internal consistency. The necessary analyses were conducted on data from 1,097 U3A students – previously involved in the EFA (n = 660) and CFA samples (n = 437).
Given the instrument’s multidimensionality and the conservative nature of the estimates, Cronbach’s α was used (Green and Yang Reference Green and Yang2015). For the entire inventory, Cronbach’s α was 0.943 (95% CI: 0.935–0.950). Similarly high values were obtained for each factor: DEV 0.844 (95% CI: 0.824–0.863), SOC 0.817 (95% CI: 0.795–0.839), AGE 0.853 (95% CI: 0.837–0.869) and HEA 0.839 (95% CI: 0.819–0.859). Removing any item did not improve the reliability of the entire scale or its subscales.
Additional analyses using an alternative measure – McDonald’s ω – yielded almost identical reliability estimates, with differences for particular components and the overall instrument ranging from 0.00 to 0.02. As expected, ω estimates were slightly higher.
The NFI-30: item discriminative power
Item discriminative power (Pearson’s r between each item and the total score without that item) was also tested (De Gruijter and Leo Reference De Gruijter and Leo2007). Data from 1,097 U3A students were reused. For the entire tool, values ranged from 0.274 to 0.640 (mean = 0.583). Satisfactory results were found for subscales: DEV 0.516 to 0.639 (mean = 0.579), SOC 0.370 to 0.637 (mean = 0.540), AGE 0.564 to 0.667 (mean = 0.594) and HEA 0.586 to 0.650 (mean = 0.617).
In summary, the final version of the inventory (NFI-30) includes 30 items representing four categories of needs: eight each for agency (AGE), development (DEV) and social (SOC) needs, and six for health (HEA) needs. Results are expressed as average scores, calculated based on responses to items assigned to each need type according to the scoring key (see Appendix 1).
Research procedure for Study 2
Study 2 evaluated the convergent and discriminant validity of the NFI-30, and its reliability defined as temporal stability.
Method
Participants
Measurement 1 of Study 2 involved 278 participants (231 females, 38 males, 9 no data; age M = 70.874, SD = 5.559), while 54 (48 females, 6 males; age M = 70.592, SD = 5.053) took part in Measurement 2, used to assess the NFI-30’s temporal stability in a test–retest procedure. The study was conducted in 10 U3As in the Silesian and Masovian voivodeships in Poland. Participant selection matched that of Study 1.
Materials and instruments
The NFI-30
Older adults’ needs fulfilled during U3A classes were assessed using the NFI-30. Items were rated on a five-point scale: 1 – definitely not, 2 – rather not, 3 – I don’t know, 4 – rather yes, 5 – definitely yes. The tool’s reliability in Measurements 1 and 2 was 0.866 and 0.903 (AGE), 0.790 and 0.761 (DEV), 0.906 and 0.818 (SOC) and 0.917 and 0.904 (HEA), respectively.
Achievement Motivation Inventory
To assess convergent validity of the NFI-30, the Achievement Motivation Inventory (AMI; Schuler et al. Reference Schuler, Thornton, Frintrup and Mueller-Hanson2004) was used. The Eagerness to Learn (e.g. item 14: ‘When I learn something new, I try to remember as much of it as possible’) and Independence (e.g. item 49: ‘I like to be responsible for what I do’) components of AMI were selected, as they were considered to represent needs analogous to those measured by the NFI-30: DEV (e.g. item 13: ‘Classes at U3A … inspire pursuit of deeper knowledge’) and AGE (e.g. item 3: ‘Classes at U3A … help me solve my problems’) factors, respectively.
Multidimensional Self-Esteem Inventory
Convergent validity was also tested using the Multidimensional Self-Esteem Inventory (MSEI; O’Brian and Epstein Reference O’Brian and Epstein1988). It was assumed that the content of the Likeability (e.g. item 6: ‘I am generally liked and popular’) and Personal Power (e.g. item 13: ‘I almost always feel healthy and in good physical condition’) components largely corresponds to the content of the SOC (e.g. item 6: ‘Classes at U3A … allow me to be around people’) and HEA (e.g. item 8: ‘Classes at U3A … improve physical fitness’) dimensions of the NFI-30, respectively.
Need for Closure Scale
To assess discriminant validity, a short version of the Need for Closure Scale (NFCSsv; Kossowska et al. Reference Kossowska, Hanusz and Trejtowicz2012) was used. The Closed-Mindedness (e.g. item 31: ‘I usually focus on one possible solution to a problem I face’) and Avoidance of Ambiguity (e.g. item 3: ‘I avoid ambiguous situations’) components of this instrument represent cognitive-motivational orientations opposite to those reflected in the NFI-30 content, characteristic of individuals open to change and development within self-constructed educational pathways.
Ten-Item Personality Inventory
Discriminant validity was also assessed using the Ten-Item Personality Inventory (TIPI-PL; Gosling et al. Reference Gosling, Rentfrow and Swann2003). Although consisting of only ten items, the questionnaire demonstrates satisfactory psychometric properties and provides valuable insight into personality structure (Extraversion, Neuroticism, Conscientiousness, Agreeableness, Openness to Experience), while not imposing excessive cognitive or physical burden on respondents completing a lengthy questionnaire.
It was assumed that needs fulfilled by older adults during U3A activities – relatively universal in liberal post-industrial societies (Formosa Reference Formosa2014, Reference Formosa2019) – should not show strong correlations with Big Five traits or, if correlated, should do so more weakly than with needs/motivations assessed using AMI and MSEI.
Research Procedure
Study 2 followed the same procedure as Study 1 (see “Research procedure for Study 1”). Measurement 1 was conducted at the end of February 2025, and Measurement 2 took place eight weeks later.
Results
The NFI-30: convergent validity
Given the conceptual and functional similarities, positive correlations were expected (Canivez Reference Canivez2025) between the NFI-30 scores and the AMI/MSEI scales, particularly for the following pairs: DEV and Eagerness to Learn, AGE and Independence, SOC and Likeability, and HEA and Personal Power.
The predictions were confirmed (see Table 3). The NFI-30 scores correlated significantly and positively with AMI and MSEI measures, and their values ranged from 0.287 (SOC – Likeability) to 0.504 (DEV – Eagerness to Learn).
Convergent and divergent validity, and temporal stability of the NFI-30
Table 3 Long description
Correlation matrix reports associations among four need domains (DEV, SOC, AGE, HEA), six related traits (EL, Likeability, Independence, Personal Power, Closed Mindedness, Avoidance of Ambiguity), and five Big Five traits. Need domains show strong positive links: DEV with SOC 0.741, DEV with AGE 0.779, and HEA with AGE 0.617; HEA also relates to DEV 0.660 and SOC 0.582. Eagerness to Learn and Likeability are moderately related to the need domains, for example EL with AGE 0.532 and Likeability with AGE 0.580. Independence and Personal Power show smaller to moderate positive correlations with needs, such as Independence with DEV 0.382 and Personal Power with HEA 0.386. Closed Mindedness has weak or negative relationships across the matrix, including with Avoidance of Ambiguity at minus 0.141 and near zero with the need domains. Big Five traits show moderate interrelations, notably Agreeableness with Conscientiousness 0.462 and Extraversion with Neuroticism 0.449, and Extraversion relates moderately to Agreeableness 0.366. Most reported correlations are statistically significant, but the matrix is lower triangular and some diagonal entries are not provided here, so interpret patterns rather than missing cells.
Notes: DEV – development, SOC – social, AGE – agency, HEA – health, EL – Eagerness to Learn, Lik – Likeability, Ind – Independence, PP – Personal Power, CM – Closed-Mindedness, AA – Avoidance of Ambiguity, Ext – Extraversion, Agr – Agreeableness, Con – Conscientiousness, Ner – Neuroticism, OE – Openness to Experience.
Temporal stability coefficients for the NFI-30 need categories are presented in bold on the diagonal of the matrix.
Significance levels: * p < 0.05, ** p < 0.01.
The NFI-30: discriminant validity
No similar associations were expected for Cognitive Closure dimensions. Since participation in U3A is voluntary and typically driven by internal motivation (e.g. desire to learn, meet new people), students should display traits opposite or only weakly related to those measured by the NFCSsv (Bowling Reference Bowling2017).
Furthermore, positive – though weaker than for AMI and MSEI – correlations with the NFI-30 dimensions were hypothesized for personality traits (Bowling Reference Bowling2017): extraversion, conscientiousness, agreeableness and openness to experience (e.g. I attend U3A classes regularly because I enjoy spending time constructively in good company) and negative correlations for neuroticism (unless U3A classes serve as a way to cope with negative emotions).
Correlations between the NFI-30 components and Closed-Mindedness (see Table 3) were insignificant (ranging from −0.007 to −0.110). Associations with Avoidance of Ambiguity were significant but weaker than those with AMI and MSEI (from 0.131 to 0.230).
Similarly, weak but positive associations emerged for personality traits. The average DEV–Big Five correlation was r = 0.160; for the remaining the NFI-30 components: SOC = 0.148, AGE = 0.115, HEA = 0.134. Finally, openness to experience correlated significantly only with the HEA factor, while Neuroticism – contrary to expectations – was positively associated with all the NFI-30 components.
The NFI-30: reliability (Temporal Stability)
The reliability of the NFI-30 in terms of temporal stability was tested using a test–retest procedure. The levels of fulfilment of the identified need categories were measured twice, in February and April 2025, in the same group of U3A students (n = 54). Scores from Measurements 1 and 2 for each need type were significantly correlated. Their values ranged from 0.511 for SOC to 0.671 for HEA, indicating that the NFI-30 demonstrates satisfactory temporal stability (Canivez Reference Canivez2025).
Gender, age and the NFI-30 results of older adults
We examined whether basic demographic variables – gender and age – were related to the NFI-30 scores. Gender did not significantly differentiate levels of need fulfilment, in the data used both in Study 1 (n = 660) and in Study 2 (n = 437). Independent samples t-tests ranged from 0.826 to 1.444 and from 0.958 to 1.072, respectively (all p > 0.05).
Similarly, the age of U3A students was not significantly associated with any NFI-30 dimensions, in both Study 1 (n = 660) and Study 2 (n = 437). Correlation coefficients ranged from −0.010 to 0.031 and from 0.041 to 0.081, respectively (all p > 0.05).
The NFI-30: provisional norms
Finally, based on data from 1,097 participants, provisional norms for each part of the NFI-30 were established. Given the left-skewed distributions of the variables, percentile ranks were used (Cohen et al. Reference Cohen, Schneider and Tobin2025). Low scores were defined as below the 25th percentile, average between the 26th and 75th, and high above the 75th percentile. Because demographic variables did not differentiate the NFI-30 scores, common norms were proposed for both gender and age groups (see Table 4).
Provisional norms of the NFI-30 scores
Table 4 Long description
Provisional score bands classify NFI-30 need scores into Low, Medium, and High based on raw score ranges for Social, Development, Agency, Health, and Global needs. For Social, Low runs from 0.00 to 3.75, Medium from 3.76 to 4.62, and High from 4.63 to 5.00. Development uses 0.00 to 3.87 for Low, 3.88 to 4.75 for Medium, and 4.76 to 5.00 for High. Agency uses 0.00 to 3.75 for Low, 3.76 to 4.50 for Medium, and 4.51 to 5.00 for High. Health uses 0.00 to 3.67 for Low, 3.68 to 4.67 for Medium, and 4.68 to 5.00 for High. Global uses 0.00 to 3.70 for Low, 3.71 to 4.52 for Medium, and 4.53 to 5.00 for High. Across needs, the Low-to-Medium cutoff falls between about 3.67 and 3.88, while the Medium-to-High cutoff falls between about 4.50 and 4.76, meaning High requires a relatively high raw score near the top of the scale. These are provisional norms, so cut points may vary with different samples or future updates.
Discussion
The NFI-30 demonstrates satisfactory psychometric properties. Considering reliability – internal consistency – it was comparable to other self-report measures assessing similar constructs. For instance, the TPS, which assesses self-efficacy (Chomczyńska-Rubacha and Rubacha Reference Chomczyńska-Rubacha and Rubacha2013), yielded a Cronbach’s α of 0.87. For the AMI, measuring achievement motivation (Schuler et al. Reference Schuler, Thornton, Frintrup and Mueller-Hanson2004), reliability values in adult populations range from 0.68 to 0.82, while for the MSEI (O’Brian and Epstein Reference O’Brian and Epstein1988), assessing self-esteem components, they span from 0.70 to 0.90.
The temporal stability of the NFI-30 was also verified. The obtained intercorrelation (r = 0.56) aligns with results for comparable tools, for example 0.69 for the Positive Orientation Scale (Łaguna et al. Reference Łaguna, Oleś and Filipiuk2011), 0.76 for the Self-Ruminating Scale (Baryła and Wojciszke Reference Baryła and Wojciszke2005) and 0.31–0.53 for the Emotion Scales (Wojciszke and Baryła Reference Wojciszke and Baryła2005).
Moreover, the theoretical validity of the instrument was confirmed, encompassing its factorial, convergent and discriminant validity. Adopted factor solutions – first for the NFI-72 and subsequently for the NFI-30, final version of the developed inventory – are largely consistent with categories that emerged from qualitative data. Thus, they reflect activities that are cognitively accessible (Hall Reference Hall2020) and/or currently undertaken by older adults in U3A classes, arising from their specific needs – social, developmental, health-related and so on.
Extracted components of the NFI-30 are also consistent with need classifications reported in literature. For example, conceptualizations cited in Introduction – Relatedness and Growth (Alderfer Reference Alderfer1972) or Coping, Contributive and Influence needs (McClusky 1963, cited in Merriam Reference Merriam1987) – correspond in content and function to the NFI-30 components SOC, DEV, HEA and AGE, respectively.
The needs identified using the NFI-30 are also largely consistent with their conceptualization within self-determination theory (Ryan and Deci Reference Ryan and Deci2000). The core needs – autonomy, competence and relatedness – correspond to AGE, DEV and SOC. One may view HEA as functionally related to autonomy, as the capacity to make independent decisions and shape one’s physical and social environment in line with personal values depends on physical and mental health. Participation in U3A activities, for example, presupposes that individuals meet HEA to a degree that enables attendance and active engagement.
In this study, the transcendental need proposed by McClusky was not identified. Although important for wellbeing in later life, U3As appear to offer limited scope for its fulfilment, as it primarily involves individual, reflective engagement with existential concerns.
Overall, the correspondence between the categories of needs fulfilled as identified using the NFI-30 and those proposed within the influential conceptualizations cited earlier is not complete. This is understandable, given that U3A students constitute a subgroup of the broader older adult population, and need categories addressed in U3A classes represent a subset of needs relevant to older adults more generally. Therefore, the NFI-30 components, developed via an inductive–deductive procedure (i.e. initial extraction of categories from qualitative data followed by evaluation of their suitability for questionnaire measurement), cannot fully align with classifications proposed for the broader older adult population, as the analysed variance differs somewhat.
Originally separate needs of adaptation and control, analysed within the NFI-72, were combined into a single ‘agency’ factor proposed in the NFI-30. This decision has theoretical underpinning. Agency can be understood as capacity to exert influence over one’s own life and social and physical environment (Horvath Reference Horvath1998). As such, it encompasses both control over life and environment and adaptation to their demands, for example ability to use a computer efficiently when making online bank transfers. This conceptualization aligns with recognized definitions of agency; for instance, according to the APA (2026) Dictionary of Psychology (https://dictionary.apa.org/agency), agency is the ‘state of being active, usually in service of a goal, or of having power and capability to produce effect or exert influence’.
Similarly, self-development and cognitive needs, assessed using the NFI-72, were combined into a single DEV factor in the NFI-30 inventory. This decision may be justified as follows: effective fulfilment of cognitive needs can be viewed as a crucial epistemic component of self-development. From this perspective, the desire to understand the world and one’s place within it stimulates learning, self-awareness and critical reflection on one’s own life. The ability to satisfy cognitive needs derives from the capacity to understand oneself, the physical and social environment, one’s limitations and one’s potential (Erikson Reference Erikson1980; Ryan and Deci Reference Ryan and Deci2017).
These assumptions are also supported by empirical findings. Grass et al. (Reference Grass, Dörendahl, Losereit, Greiff and Strobel2023) showed that active seeking of intellectual stimulation is strongly associated with positive affect and higher job and academic satisfaction. Here, cognitive activity functions as a resource for personal fulfilment. Lavrijsen et al. (Reference Lavrijsen, Aerts, Preckel, Ramos and Verschueren2024) conceptualized cognitive need as a crucial feature of individuals’ ‘intellectual investment’. The authors argued that cognitive effort shapes how individuals allocate their potential across the lifespan, directly influencing trajectories of educational and personal self-development.
The proposed four-factor structure of the NFI-30 and its corresponding items aligned well with data from 1,097 U3A participants. Importantly, categories of fulfilled needs were not derived solely from the literature and imposed on respondents, but emerged inductively from respondents’ spontaneous statements. This suggests that the proposed inventory reflects conceptual categories genuinely used by older adults to describe needs met during U3A classes. In other words, the NFI-30 shows satisfactory ecological validity (Wegener and Blankenship Reference Wegener, Blankenship, Baumeister and Vohs2007).
Nevertheless, it should be clearly noted that the proposed introductory wording for the NFI-30 item (‘Activities in U3A…’, see Appendix 1) implies that the tool measures perceived institutional contribution to need satisfaction rather than an objective, that is, an institution-independent level of need fulfilment among individuals of a specific age group. This has important implications for generalizability and for comparisons across institutions involved in educating and stimulating older adults.
Triangulation of approaches and sources suggests that both studies succeeded in identifying key needs of older adults, which can be considered relatively universal motivators of activity co-occurring with processes and phenomena typical in older adult biographies, such as loneliness caused by death of partner or peers. Older adults can effectively satisfy these needs by taking advantage of rich U3A offerings, as U3A effectively addresses needs of individuals in this age category. In this light, needs addressed in U3A can be regarded as intra-psychic structures of older adults that encourage engagement in various educational and social activities, which largely define quality of life (Stuart-Hamilton Reference Stuart-Hamilton2012; Formosa Reference Formosa2014).
The NFI-30 results correlated positively with measures of conceptually and functionally similar constructs (motivation and self-esteem components), confirming the instrument’s convergent validity. In turn, correlations with conceptually distant constructs (e.g. need for cognitive closure, personality traits) were weak or non-significant, supporting its divergent validity.
An exception was the significant correlations between the neuroticism trait and all components of the NFI-30. This unexpected finding may be explained by the possibility that heightened emotional sensitivity among older adults increased their need for mood-enhancing activities in the friendly atmosphere of U3A classes. This interpretation is supported by the significant correlation between Neuroticism and SOC, the strongest among those observed between personality traits and the NFI-30 components (see Table 4).
Neurotic individuals – that is, those with elevated emotional tension – may actively seek accepting environments and place higher value on opportunities to satisfy multiple needs within them. Furthermore, neuroticism is linked to lower satisfaction in social relationships (Verduyn and Brans Reference Verduyn and Brans2012; Nilsen et al. Reference Nilsen, Bang and Røysamb2024), reflecting tendencies to experience negative emotions and engage in rumination, which can reduce perceived fulfilment of needs, such as belonging and emotional responsiveness, in domestic contexts. Accordingly, neurotic individuals may particularly appreciate safe settings provided by U3A classes and opportunities to satisfy their needs within these environments.
More intensive fulfilment of needs for social approval, personal growth or agency may also serve as a defensive mechanism against strong anxiety. Such tension can be effectively alleviated through participation in U3A classes, providing accepting peer relationships and skills for coping in a stressful, rapidly changing world (Apostolov and Geldenhuys Reference Apostolov and Geldenhuys2022). In summary, when U3A provides a supportive environment fostering emotional expression and access to social support, and when instrumental competencies developed therein constitute practical tools for influencing one’s life and social and physical environment (Noble et al. Reference Noble, Medin, Quail, Young and Carter2021), neuroticism may be positively associated with SOC, DEV, AGE and HEA.
Limitations
The proposed instrument has certain limitations worth noting. First, the NFI-30 is tailored to a specific group of older adults – U3A participants – who may differ from the general older population, for example in terms of curiosity, or a stronger need for social, cognitive and physical stimulation.
It appears feasible to develop a more universal tool by modifying the introductory phrase of the items from ‘During U3A classes …’ to a more general form, such as ‘On a daily basis …’ or ‘Usually …’. Work on this is underway, and the data collected will serve to evaluate the psychometric properties of the adapted scale for the broader older adults population.
Second, the content and predictive validity of the NFI-30 have not yet been assessed. This step seems valuable given the intended practical relevance of assessing older adults’ need fulfilment and the tools designed to measure it. In terms of content validity, we consider that the inductive–deductive procedure enabled us to identify need types and corresponding items that adequately represent cognitive, social, health and developmental activities of U3A participants.
With regard to predictive and diagnostic validity, work is underway to develop a more universal instrument for assessing needs satisfied by older adults in general. As part of this research, we compare U3A students with their less active peers, expecting clear differences in favour of U3A students.
In the version of the instrument for older adults in general, the introductory instruction reads: ‘When describing myself, I can say that …’. To date, the sample includes n = 159 inactive older adults, with data collection still ongoing. For this article, we compared responses of inactive older adults with those of U3A students. Consistent with the directional hypothesis, all four need categories were higher among U3A students than inactive peers: for DEV (MU3A = 4.235; SD = 0.630 vs MnAct = 3.882; SD = 0.719; t(944) = 6.042; p < 0.001; Cohen’s d = 0.548), for SOC (MU3A = 4.156; SD = 0.658 vs MnAct = 4.046; SD = 0.692; t(927) = 1.909; p = 0.028; Cohen’s d = 0.180), for AGE (MU3A = 3.975; SD = 0.748 vs MnAct = 3.645; SD = 0.727; t(947) = 4.792; p < 0.001; Cohen’s d = 0.443) and for HEA (MU3A = 4.120; SD = 0.759 vs MnAct = 3.777; SD = 0.882; t(980) = 4.977; p < 0.001; Cohen’s d = 0.442).
Third, the reliability and accuracy of the NFI-30 were assessed using data from samples that were not balanced in terms of contextual variables relevant to recognizing and fulfilling various needs – for example gender, education level or marital status. Higher cognitive and social activity is often more typical of women with higher education than men with vocational training (Kan et al. Reference Kan, Zhou, Negraia, Kolpashnikova, Hertog, Yoda and Jun2021). Although the current study found no gender differences in the fulfilment of the four need categories, future validation efforts should consider such potential differences and strive for balanced samples.
Furthermore, in terms of convergent validity, the NFI-30 yielded a somewhat unexpected result, given the low correlation (r = 0.278) between SOC (measured by the NFI-30) and Likeability (measured by the MSEI). This finding, inconsistent with the prior assumptions, suggests that results obtained using the instrument should be interpreted with caution, particularly with respect to the SOC dimension and its construct adequacy.
Fourth, the proposed norms should be regarded as provisional and interpreted with appropriate caution. They were derived from data obtained in a single study and, as such, require validation using evidence from independent replication samples. Norms based on a single sample may lack robustness and may reflect idiosyncratic characteristics of the initial validation sample, even when the sample size is relatively large (n = 1,097). For instance, the observed very high mean levels of need fulfilment among U3A students may be indicative of a ceiling effect. Ongoing research by the instrument’s authors is currently addressing the diagnostic and prognostic validity of the NFI-30.
Despite these limitations, the NFI-30 is a valuable, multidimensional tool for quickly assessing the level of fulfilment of key needs among older adults. Based on reliable data, it enables the design of effective preventive and educational programmes aimed at enhancing quality of life in this demographic group (Noble et al. Reference Noble, Medin, Quail, Young and Carter2021).
Conclusions
A reliable diagnosis of how older adults’ needs are met in U3A classes supports a better understanding of their functioning in various contexts – both educational and non-educational (e.g. family, peers). As a result, first, U3A organizers can make evidence-based evaluations of the educational offer and adapt it to specific groups. Institutions bringing together multiple U3A units enable the use of the NFI-30 instrument to diagnose their capacities, considering existing infrastructure and organizational resources. Such analyses may serve as a basis for designing tailored educational programmes within individual units. Second, policymakers can identify areas where the U3A offer falls short and propose improvements. Based on the diagnostic findings, they may identify neglected areas that could be addressed by U3As or other organizations, for example through commissioning programmes via competitive grants or by employing relevant specialists within existing educational institutions. And third, researchers – after appropriate linguistic adaptation of the NFI-30 – can assess older adults’ needs in their countries and examine links with other variables relevant to quality of life, such as meaning, optimism or achievement motivation.
In summary, the use of the NFI-30 may contribute primarily to enhancing the quality of life among older adults who are already engaged in educational activities. Further adaptation of the instrument to other target groups may facilitate its broader application, including among older adults who are socially excluded or otherwise disadvantaged.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0144686X26100889
Financial support
This article did not receive funding.
Competing interests
The authors declare that there is no conflict of interest.
Ethical standards
The study protocol was approved by the Ethics Committee for Research at WSB University, Poland, approval number AWSB/KE/1/2025. Datasets underlying the presented study are available here: Trusz S, Fabiś A, Duraj A and John B (2025) ‘Datasets and materials underlying the Study on Needs Inventory of Third Age University Students: Tool Development and Psychometric Properties’, https://doi.org/10.7910/DVN/DWCBOE, Harvard Dataverse, V2.
