Unleashing the Value of Lifelong Learning Institutes: Research and Practice Insights From a National Survey of Osher Lifelong Learning Institutes

Abstract

Lifelong learning as a field has leapt forward, but work remains to inform the structuring and practice of lifelong learning institutes. This study furthers the field of institution-based lifelong learning by utilizing cross-institutional research regarding the value of lifelong learning to older adults. A content analysis coding approach is undertaken to categorize and organize 4,400 lifelong learners’ responses to the question: “What about your Osher Institute has been valuable/important to you?” Responses came from 12 institutes across the United States. The organization and interpretation of the themes fell along four dimensions: (a) learning experience, (b) community environment, (c) learning quality, and (d) learning access. Subthemes within each of the four broad themes are identified and discussed. Slight demographic differences regarding codes assigned are highlighted. Institutional differences were more notable demonstrating that the structures and practices of institutes warrant more research.

Keywords

lifelong learning research lifelong learning practice adult education older adults

Background and Purpose

What is the value of lifelong learning? Such a question has implications for lifelong learning institutes as well as implications for current and potential lifelong learners (Boulton-Lewis, 2010). While the institutional, and often quantitative, indicators of value have been explored in the literature, the qualitative value of lifelong learning at various points of the human lifespan deserves further exploration (Talmage, Lacher, Pstross, Knopf, & Burkhart, 2015). This study explores the value of lifelong learning to older adults (i.e., aged 50+ years) and how the value of lifelong learning is perceived and experienced across selected U.S. lifelong learning institutes.

The growth of the older adult population couples with a growth in the number of lifelong learning providers for older adults (Formosa, 2014), such as the Osher Lifelong Learning Institutes (OLLIs)¹ in the United States (Hansen, Brady, & Thaxton, 2016), Elderhostel programs (Cain, 2008), and Universities of the Third Age (Formosa, 2014). U.S. lifelong learning institutes are often grounded in holistic and positive approaches to aging that leverage older adult strengths (e.g., Talmage et al., 2015), rather that framing the older adult experience in terms of decline or and deficiencies (Tornstam, 2011). These institutes do not simply construe lifelong learning as a tool for physiological or social stimulation (Boulton-Lewis, 2010), but also seek to provide transformative moments through learning and community engagement (Pstross, Talmage, Peterson, & Knopf, 2017). They seek to highlight and celebrate their members’ broad experiences and uniqueness (Hansen et al., 2016; Talmage et al., 2015); however, diversity in demographics remains a challenge (Hansen et al., 2016; Lee, 2016).

Lifelong learning is a historical cornerstone of adult education and the learning society (Jakobi, 2012; Katz et al., 2008). While lifelong learning has been politically leveraged for economic development, these views are reductionist (Jarvis, 2000). Instead, this study looks at the perceived value of lifelong learning to older adults outside of traditional skill-building learning programs. It also explores value from the learners’ perspectives, not provision or policy. This exploration directly exploits the gaps in the literature regarding a lack of large scale, national studies of the value of lifelong learning from lifelong learners’ points of view.

Exploring the Value of OLLIs

Bernard Osher began funding the OLLI network of U.S. university/college-affiliated programs in 2001 through The Bernard Osher Foundation. Now, 121 OLLIs engage over 160,000 total members/learners collectively. The OLLI network includes a National Resource Center for Osher Lifelong Learning Institutes (NRC). The Osher NRC hosted at Northwestern University aims to connect, collaborate, and consult with OLLIs to help them thrive. This study’s data come from Osher NRC’s 2016 national demographic survey conducted across 12 OLLIS.

OLLIs engage learners in college/university-level learning experiences and provide them with communities of other lifelong learners (Hansen et al., 2016; Talmage et al., 2015). These experiences encompass a diverse assortment of college-level learning events and courses tailored for learners aged 50 years and older. OLLIs aim for learners to be “in charge of their own learning experiences . . . what to learn, how to learn, with whom to learn, what goals to set for learning, and what value they get from learning” (Talmage et al., 2015, pp. 233-234).

The Value and Value Components of Lifelong Learning

Lifelong learning provision continues to expand as population growth among older adults burgeons (Formosa, 2014). Lifelong learning is both a gift and a service as it promotes higher levels of well-being in older adulthood by addressing the changing demands and interests of older adults (Boulton-Lewis, 2010; Jenkins, 2011). Unfortunately, the value of lifelong learning may not be fully recognized by older adults (Hebestreit, 2009). Older adults differ in what they value, but they may not be given an opportunity to share what they value apart from whether they enroll in particular learning offerings (e.g., Talmage et al., 2015).

For this study, four core components of the value of lifelong learning to learners were identified based on a preliminary reading of the study data (discussed later). These components were investigated in the literature to identify subthemes to be used for content analysis coding and interpreting the study’s findings. The four core components of value are as follows: (a) learning experience, (b) community environment, (c) learning quality, and (d) learning access. These components constitute neither a complete nor an all-inclusive model, but are utilized to prompt in-depth discussion about the dimensionality of the value of lifelong learning.

Learning experience

While some have argued that older adults want to choose how to learn (e.g., Talmage et al., 2015), more often lifelong learning institutes make choices on the learners’ behalf—hoping they will satisfy the expectations and interests of the older adults they serve (Clark, Fochs Heller, Rafman, & Walker, 1997). Positive reactions to the topics learned are important (Weinstein, 2004). For instance, individuals may indicate that they like what they have learned or feel good learning about particular topics or the learning experience in general. While positive feelings are important to help individuals retain what they learn (Kirkpatrick, 1994), the primary source of value regarding the learning experience likely comes from the insights gained from learning. These insights may also in turn facilitate the experience of joy for lifelong learners (Lamb & Brady, 2005; Weinstein, 2004). Considering both positive reactions and learning insights, the following question is asked:

Research Question 1: What is the value of the learning experience to older adults considering both positive reactions and learning insights?

Community environment

Social motives are important to older adult learners but secondary to cognitive desires to learn (Kim & Merriam, 2004); however, the claim of social motives falling secondary to cognitive motives has been challenged (e.g., Clark et al., 1997). It remains important to identify the unique and distinct value of the community environment and its facets in lifelong learning, because social motives often appear in positive stories from lifelong learners (Illeris, 2003). Lifelong learning directors are aware of the importance of socially focused activities for their programs (Brady, Cardale, & Neidy, 2013).

Lifelong learning programs provide older adults the opportunity to socialize with others, which can occur before, during, and after class or special events (Golding, 2011). If older adults enjoy socializing, then perhaps, they also value the ability and experience of learning with others, often termed collaborative learning (e.g., Oddi, 1987). In this study, collaborative learning is viewed as collaborating with other learners in the classroom, not necessarily cooperating with the instructor on what and how to learn (Marsden, 2011).

Deeper than collaboration, community building can lead to the application of lessons outside of the classroom and local community engagement, which may be of great value to lifelong learners (Merriam & Kee, 2014; Pstross et al., 2017). Community-building activities include interactions with others that foster a shared sense of community (McDonough & Davitt, 2011). An interlinking order might be considered, such that socializing serves as a first step to community-building or collaborative learning, or the links may also reciprocate. Considering the community environment, the following is asked:

Research Question 2: What is the value of the community environment to older adults considering socializing or social activities, learning with others, and community-building activities?

Learning quality

Older adults value high-quality learning experiences, but which aspects of that value are most important? Research has identified some key facets, such as quality instruction and instructors (Lamb & Brady, 2005), quality programming and staff (Dauenhauer, Steitz, & Cochran, 2016), course variety (Talmage et al., 2015), and off-campus or extracurricular offerings and experiences (Talmage et al., 2015). Considering these facets, it is essential to ask:

Research Question 3: What is the value of the learning quality to older adults considering quality instruction and instructors, quality programming, variety of course offerings, and off-campus offerings and experiences?

Learning access

Learning must be accessible to older adults for valued to be realized (Chen, Kim, Moon, & Merriam, 2008; Lamb & Brady, 2005; Merriam & Kee, 2014), but how much is access valued by older adults? Lifelong learning institutes and their programs aim to be physically and financially accessible to older adults (Talmage, Mark, Slowey, & Knopf, 2016). For example, OLLI at ASU validates parking for learners and allows students to register online, by phone, or by mail. It also provides YouTube videos explaining how to register online for courses. Furthermore, courses generally have no tests or homework, and courses are shorter than traditional college terms or semesters, making the learning process easier. If access is valued, the particular access aspects must be identified. Thus, the following is asked:

Research Question 3: What is the value of access to learning for older adults considering both the access to learning experiences and the ease of the learning process?

Method

The current study utilizes a holistic approach allowing older adults engaged in lifelong learning to identify the value they find in lifelong learning. Mixed and multiple research methods honor the expressions of older adults (Talmage et al., 2015). Four research questions are examined using an exploratory quantitative approach drawing on content analysis coding methods. The content analysis coding methods analyze data from an open-ended question accompanying other quantitative demographic information ascertained. The content analysis coding method transforms qualitative responses into quantitative data (Hsieh & Shannon, 2005).

Data Source and Information

The data comes from the Osher NRC’s 2016 national demographic survey, which was conducted across 12 lifelong learning institutes from the OLLI network. Institutes were selected by Osher NRC staff to attain geographic and demographic diversity. Enrollees of the 12 institutes received a questionnaire. However, 96% of enrollees completed the questionnaire online; those without e-mail access could complete a paper survey entered online later by NRC staff. Indeed, 5,509 total responses were collected. Specifically, this study focuses on a single qualitative measure: What about your Osher Institute has been valuable/important to you? A total of 4,400 individuals responded to this qualitative question (see Table 1).

Table 1.

Location and Item Response Rates.

Location	Response frequency	Percentage of total sample	Number surveyed	Qualitative item response rate, %
Boise State	433	9.84	538	80.48
Furman University	542	12.32	635	85.35
University of Connecticut	174	3.96	207	84.06
University of Southern Maine	648	14.73	825	78.55
University of Delaware	1,013	23.02	1,227	82.56
San Francisco State University	212	4.89	266	79.70
University of Kansas	518	11.78	761	68.07
Hampton University	83	1.89	118	70.34
University of California, Irvine	163	3.71	198	82.32
Northwestern University	360	8.18	413	87.17
Colorado State University	119	2.71	139	85.61
University of Miami	135	3.07	182	74.18
Total	4,400	100.00	5,509	79.87

The analysis began by content coding of the responses—translating qualitative content from the respondents into quantitative data—but then returned to the qualitative information to form a platform for formulating study inferences. Demographic information about the sample is presented in Table 2. As part of the analysis, demographic measures were used to investigate consistency across responses using chi-square tests of independence.

Table 2.

Demographics of Item Respondents.

Participant characteristics	Frequency	Percentage
Gender
Female	2,986	67.86
Male	1,401	31.84
No answer	13	0.30
Marital status
Married/partnered	2,803	63.71
Single	880	20.00
Widow(er)	698	15.86
No answer	19	0.43
Age group, years
<55	28	0.64
55-59	97	2.21
60-64	468	10.64
65-69	1,188	27.00
70-74	1,266	28.77
75-79	773	17.57
80-84	370	8.41
85+	201	4.57
No answer	9	0.21
Education level
Grade school or some high school	5	0.11
High school diploma or GED	93	2.11
Associate’s degree or equivalent	249	5.66
Bachelor’s degree or equivalent	1,262	28.68
Master’s degree or equivalent	1,851	42.07
Doctoral or other professional degree or equivalent	757	17.21
Other	169	3.84
No answer	14	0.32

In general, the majority of persons in the sample were between 60 and 79 years of age, were women, were married/partnered, held at least a bachelor’s degree, and self-identified as of White/Caucasian race or ethnicity. Only 5.25% identified as non-White/Caucasian, so race/ethnicity was excluded for comparative analyses because of low sample size. This sample’s composition is not uncommon for most university-centered lifelong learner populations; thus, the challenge remains to attain diversity across samples and programs (e.g., Lee, 2016).

Content Analysis Coding and Rigor

Code development

Four researchers, independent of the coders, crafted an a priori coding matrix, which centered on the research questions driving this article. These four researchers were affiliated with the Osher NRC and three OLLI locations (two surveyed for this study) driving this article. The five aforementioned themes identified by the researchers from a preliminary reading of the study data (also noted earlier in the introduction) were as follows: (a) learning experience, (b) community environment, (c) learning quality, (d) learning access, and (e) other.

The researchers then independently examined the qualitative data to validate and expand the a priori structure based on their informed intuition and guided by themes found in the literature. Subsequently, through a series of interactive video conference calls, the four researchers reached consensus on a 12-point a priori coding structure that was judged capable of adequately capturing the diversity of responses in the data set, relative the literature that gave rise to the study’s research questions.

Next, two of the four researchers, who were not part of the original survey questionnaire development but who were subject matter experts in lifelong learning and lifelong learning institute research, undertook a high-level examination regarding whether these themes and subthemes were prevalent across the 12 locations. Deeming themes and subthemes common across locations, the two researchers then presented them to the initial team of four for final agreement on the 12 codes (see Appendix A), a process consistent with recommended methods for rigor such as peer debriefing (Talmage et al., 2015).

Coding and coders

Two trained researchers separate from the four aforementioned researchers who developed the coding matrix were enlisted to content code the 4,400 open-ended responses. These coders had no association with the OLLI network, but were trained by one of the researchers who developed the codebook and its 12 subthemes. During the training, the two coders read and reread a selection of responses from one location in the presence of the researcher who trained them and discussed how they felt the response should be coded. This form of interactive feedback training regarding interpretation of the codes was conducted because of its capacity to increase interrater agreement (Pulakos, 1984). Two coders were determined to be sufficient because of the straightforward descriptions of the codes and the high volume of responses needing to be coded (Talmage et al., 2015).

Separately, the two coders read and reread each response to see if it fit one of the subthemes. Using Microsoft Excel, the coders assigned a “1” to the response in a column associated with the subtheme if the response reflected the subtheme from the codebook/training. If a response did not reflect the subtheme, the coders left the column blank; zeros in Microsoft Excel and SPSS data sets later replaced the blanks. The coders were permitted to assign multiple codes/subthemes to each response.

The interrater agreement for coding was 85.69%, reasonably sufficient (Hartmann, 1977; Stemler, 2004). It is well known and shown in this study (χ² = 346, p < .001, Φ = .232) that coders and how they code are rarely independent (Armstrong, Gosling, Weinman, & Marteau, 1997); overlaps and biases will often occur, but for this study they were low. The observed effect size (phi-correlation) investigating independence between codes assigned and raters was low, but also significant because of the large sample size. When raters did not agree that a response deserved a particular code, the code was still assigned because one person deemed the response as fitting with the code provided in the codebook. This approach appeared worthwhile despite the small-observed effect because the difference between observed and expected frequencies for which a response should not receive a code was close to zero ([f_o − f_e]/f_e = .027).

Furthermore, the coders, their trainer, and other research members team utilized reflexivity (i.e., reflecting on their roles as researchers, practitioners, and lifelong learners) and peer debriefing during the coding and analyses processes to ensure methodological rigor (Horsburgh, 2003; Lietz & Zayas, 2010). The reflexivity coupled with peer debriefing refined the final versions of codes (Appendix A) and training received by the coders, who also provided feedback on the clarity of the codes and the coding process. Notes and records of meetings were kept to review decisions made by the research team.

Finally, chi-square tests of independence and phi-correlations were used to look at relationships between codes assigned (i.e., overlaps or the tendency to multiple codes given to responses), which helped justify subtheme uniqueness within a larger theme. Chi-square tests of independence were also utilized for comparisons, because of the nominal measures utilized, to assess if the codes (i.e., the subthemes) were consistent across locations and demographics. These comparisons help build not only a case for consistency in codes and themes but also allows for future investigations of the unique differences unearthed.

Results

Prevalence of Subthemes

Frequency distributions for the 12 subthemes are presented in Table 3 and exemplary responses are provided in Appendix B. The most prevalent subtheme was insights from learning, which captured over 30% of the responses. The least prevalent subthemes were off campus offerings and experiences and ease of learning; these subthemes captured between 2% and 3% of responses. The other nine subthemes each represented between 5% and 20% of responses. A total of 8.25% of the 4,400 responses were assigned the fits no other category code.

Table 3.

Prevalence of Subthemes With Chi-Squares and Effect Sizes Highlighting Demographic Differences.

Theme and subtheme	Frequency	Percentage	Location^a	Gender^b	Marital status^c	Age group^d	Education level^e
Learning experience
Positive reactions to learning	415	9.43	χ² = 161.861, V = .192	—	—	—	—
Insights from learning	1,350	30.68	χ² = 358.840, V = .286	χ² = 12.749, Φ = −.054	—	χ² = 13.717, V = .056	χ² = 31.599, V = .085
Community environment
Socializing/social activities	878	19.96	χ² = 135.935, V = .176	χ² = 5.054, Φ = −.034	χ² = 7.077, V = .040	—	χ² = 14.645, V = .058
Learning with others activities	519	11.80	χ² = 302.888, V = .262	—	—	—	χ² = 17.691, V = .064
Community-building activities	268	6.09	χ² = 123.473, V = .168	—	—	—	—
Learning quality
Quality instruction and instructors	623	14.16	χ² = 246.284, V = .237	—	χ² = 9.827, V = .047	—	—
Quality programming	788	17.91	χ² = 151.947, V = .186	—	—	—	—
Variety of course offerings	562	12.77	χ² = 122.032, V = .167	—	—	χ² = 14.601, V = .058	—
Off-campus offerings and experiences	119	2.71	χ² = 237.853, V = .233	—	χ² = 7.137, V = .040	—	—
Learning access
Access to learning experiences	241	5.48	χ² = 73.364, V = .129	—	—	χ² = 25.278, V = .076	—
Ease of the learning process	92	2.09	χ² = 35.948, V = .090	—	—	χ² = 24.014, V = .074	—
Other
Fits no other category	363	8.25	χ² = 91.859, V = .144	χ² = 7.164, Φ = .040	—	—	χ² = 22.439, V = .072

All presented tests of independence regarding location and the subthemes are significant at the p < .001 level indicating the lack of independence. ^bGender and socializing (p < .05), gender and other (p < .01), and gender and insights from learning (p < .001) were significant indicating the lack of independence. ^cMarital status and socializing and off-campus offerings and experiences (p < .05); marital status and quality instruction (p < .01); indicating the lack of independence. ^d<55 Excluded due to group size; age groups and insights from learning and variety of course offerings (p < .05); age groups and access to and ease of learning (p < .001). ^eLess than high school excluded due to group size; education level and learning insights and other (p < .001); educational level and socializing (p < .05); education level and learning with others (p < .01) were significant indicating the lack of independence.

Efficacy of the Subthemes and Approach

All of the 4,400 qualitative responses showed capacity to reflect at least one of the 12 codes (i.e., subthemes). Some responses did receive multiple codes, which was allowed. The majority of responses (65.71%) only reflected one subtheme. Over a quarter (28.39%) of the responses reflected two subthemes. Less than 5% (4.91%) of responses reflected three subthemes. Less than 1% (0.89%) of responses reflected four subthemes. Only three responses reflected five subthemes and two responses reflected six subthemes.

Overall, the cross-correlations between the codes are not worrisome. The phi-correlations among the subthemes were low (−.233 < Φ < .140) as seen in Table 4. These low correlations demonstrate the relative uniqueness of each of the codes (i.e., subthemes). Still the cross-correlations observed suggest that codes/coding may benefit from framing more unique codes.

Table 4.

Phi-Correlations Among Subtheme Mentions.

Theme	Learning experience (A)		Community environment (B)			Learning quality (C)				Learning access (D)		Other (E)
Subtheme	Positive reactions to learning (A1)	Insights from learning (A2)	Socializing/social activities (B1)	Learning with others activities (B2)	Community-building activities (B3)	Quality instruction and instructors (C1)	Quality programming (C2)	Variety of course offerings (C3)	Off-campus offerings and experiences (C4)	Access to learning experiences (D1)	Ease of the learning process (D2)	Fits no other category (E1)
A1	—
A2	−.151***	—
B1	ns	.060***	—
B2	−.106***	−.101***	−.147***	—
B3	−.069***	ns	−.099***	−.037*	—
C1	−.095***	−.142***	−.120***	−.068***	−.068***	—
C2	−.118***	−.232***	−.058***	−.103***	−.042**	.089***	—
C3	−.107***	−.179***	−.092***	−.087***	−.066***	.139***	−.031*	—
C4	−.035*	−.059***	ns	−.048**	ns	ns	.032*	ns	—
D1	−.054***	−.102***	−.073***	−.073***	−.049**	ns	.041**	.063***	ns	—
D2	−.036*	−.039*	ns	−.044**	−.031*	ns	ns	ns	ns	ns	—
E1	−.074***	−.142***	−.133***	−.099***	−.073***	−.089***	−.106***	−.092***	−.035*	−.050**	−.044**	—

p < .05. **p < .01. ***p < .001. (All two-tailed tests).

The coders judged at total of 363 responses (8.25%) to contain content that fell outside the 11 a priori (i.e., non-Other category) codes and in the other category code/theme. Only a subset of 288 responses (6.55%) were coded as only fitting no other category and no a priori theme/subtheme; thus, 93.45% of responses received codes based on the a priori themes and subthemes. After further review, a few additional themes were noted, but the number of responses associated with them was low (less than 2% each of the total response as a maximum potential for each subtheme). Thus, no further subtheme differentiation was made. Examples were expressions of dissatisfaction with programs, expressions of satisfaction with specific courses or instructors, satisfaction with the refreshments provided, and comments about being new to programs. Some respondents commented that they valued teaching courses themselves. Overall, the other responses did not reflect content particularly relevant to lifelong learning.

Demographic and Location Differences in Subtheme Prevalence

The chi-square tests of independence investigated consistency in subthemes across locations and demographics. While Table 4 shows some demographics may not be independent of subtheme/code prevalence, the practical significance of these findings is very low. Effect sizes (i.e., Phi and Cramer’s V) were close to zero, with none exceeding ±.090.

Notable among the results of chi-square testing was the rejection of independence between location and subtheme prevalence. Markedly, the effect sizes were larger than the other demographic variables (Cramer’s V ranging from .090 to .286). Given their statistical and practical significance, a selection of institutional contexts is provided later in the discussion.

Discussion

This national study of OLLI learners highlighted the values older adults attach to experiences provided by lifelong learning institutes and programs. Prevalent value themes and subthemes exist (Figure 1) and were generated by the review of literature and coding, which demonstrated varying degrees of importance to lifelong learners. Learners’ exemplar responses are quoted throughout the discussion.

Figure 1.

Core components of the value of lifelong learning.

Research Insights

Learning experience

Insights from learning are valued the most in terms of expressed subthemes from the entire sample. This is not surprising, given the main tenet of lifelong learning is to indeed learn (Lamb & Brady, 2005; Weinstein, 2004). Older adults mentioned learning insights nearly three times as frequently as positive reactions to the learning experience. This difference leads to the question, “Which is more important: Liking the learning experience or learning something new?” This study finds that insights are more valued than reactions, consistent with previous learning models (e.g., Kirkpatrick, 1994). Still positive reactions may be important antecedents to learning insights. Future research should investigate possibile value-chains/links, as some responses contained both themes:

These classes have opened and expanded my mind to new ideas, concepts, and have enthralled me. Things have come together in my learning, which has been so exciting. The knowledge I gained is unbelievable. I look forward every semester to the new Osher course catalog. These classes keep me intellectually engaged and for me it [is] actually a source of entertainment.

Community environment

The second most frequently populated subtheme is socializing and social activities, consistent with the general revelations of past research that typically reveals that the motive for socialization is shadowed by the focus on intellectual stimulation (Kim & Merriam, 2004). The responses showcase the deeper social and communal aspects of the learning environment valued by lifelong learners; however, many responses were quite short like: “Meeting old friends and new people; or, the wonderful diverse people you meet.”

Individuals value learning with others, also called collaborative learning, more than activities focused on community-building. These findings beckon for further articulation of the community-building strategies present within OLLIs, and also, the community-building orientations of OLLI lifelong learners. Researchers have highlighted the benefits of community-building to lifelong learners (Pstross et al., 2017), but lifelong learners may be less aware of those benefits than researchers.

Some qualitative responses highlight common interests that connect collaborative learning and community building. There were responses such as “Like minds, eager to learn and discuss and Great friendships with like-minded people.” While like-mindedness was evident among the responses, some noted that they valued hearing others’ different points of view as well.

Learning quality

Quality programming, quality instruction, and course variety fall third, fourth, and fifth, respectively, among the most prevalent subthemes mentioned by the respondents. These high frequencies highlight the need for future research on the operational structures of lifelong learning institutes. Institutes can and must be structured to provide both breadth and depth in learning to constituents (Talmage et al., 2015).

Quality programming was the most prevalent learning quality subtheme, and the third most prevalent subtheme overall. While some research has examined which classes are of interest (e.g., Talmage et al., 2015), future research must examine which programming aspects are most important. While much of the research on adult education programming has treated older adults as homogeneous, diversity likely requires greater attention to provide the best programs (Chen et al., 2008). This study found quality instruction and course variety to be important, consistent with previous research (Eyler, 2002; Talmage et al., 2015). Lifelong learners care about who is teaching their courses and how courses are taught.

Off-campus offerings/experiences were substantially less prevalent, but were still mentioned. Future research might consider which offerings off-campus are best suited for the lifelong learning context. The low prevalence may relate to the structures of the institutes involved in the study; off-campus experiences varied across institutes. The theme also lacked greater specificity. Still these findings demonstrate that transformation in lifelong learning occurs primarily on-campus (Pstross et al., 2017; Talmage et al., 2016).

Learning access

Access to learning showed a lower frequency of mentions compared with other themes and subthemes. The rationale for this finding might be that learners generally expect accessibility. To occur, learning must first be accessible (Merriam & Kee, 2014). While both subthemes were mentioned less than the other three component areas, access to learning experiences was mentioned more than the ease of the learning process. Common access topics were parking, public transportation, costs, location, course availability, and convenience, which have been considered by institutes seeking to be more age friendly (Talmage et al., 2016). Future research might benefit more from considering access as expectations rather than values.

Practice Insights

This study’s findings provide lifelong learning institutes with insights regarding what learners may value most or at least value first. Lifelong learning providers must provide high-quality learning experiences where students gain insights from learning new things. Congruently, lifelong learners value high-quality instruction and programming. Specifically, this study highlights that variety of course offerings is not as highly valued as insights, instruction, or programming. Access to learning experiences may be implied, as it was not shown to be highly valued, so practitioners likely should not forgo their efforts toward greater access.

Around one in five respondents valued socializing or social activities most or at least first; however, learning with others and community-building activities were not as highly valued. This study provides lifelong learning providers with a rough estimate of how many lifelong learners value the social aspects most or first. Furthermore, the lower prevalence of learning with others and community-building activities should not be overstated to practitioners. Institutional variations in facilitating collaborative learning and community building must be considered.

This study’s institutional differences were not expected and were notable. While broad generalizations or commentary are outside this study’s purview, descriptions of three different OLLIs are offered to highlight the importance of considering institutional differences in future studies on the value of lifelong learning (Table 5). Diving into the chi-square analyses, OLLI at Boise State University appears to demonstrate greater values of insights from learning, quality instruction, quality programming, and course variety more than statistically expected. OLLI at Furman University learners value insights from learning and community building activities more than expected, but quality programming, quality instruction, and positive reactions slightly less than statistically expected. OLLI at Northwestern learners valued insights from learning more and learning quality and positive reactions less than statistically expected.

Table 5.

Selected Descriptions of Lifelong Learning Institutes.

Lifelong Learning Institute	Description	Selected membership notes
OLLI at Boise State University	OLLI at Boise State University (a public university) is located in a growing metropolitan area. The program has dedicated classrooms at its single site just off campus. This program has a slightly younger skew with the peak in age distribution of 65- to 69-year-olds, compared with 70- to 74-year-olds for the other OLLIs in 2014 and 2016 surveys. The program enjoys a strong integration with campus life and strong administrative support. Emphasis is placed on recruiting instructors from within the university’s current, adjunct, or emeritus faculty or those from other local colleges in the community. Local community experts/members are also used.Learning and quality measures are particularly emphasized; instructors are evaluated with a standardized assessment tool, with measures focusing on the expectation of outstanding presentation skills. The program provides each instructor with practice guidelines for conducting effective presentations to the OLLI audience, and staff persons are available for presentation rehearsals, as needed. The program focuses on presenting a variety of courses that are more academic than other local community programs. Members typically participate in about eight events or multiple week courses annually.	1,600 learners, 140 members volunteer
OLLI at Furman University	OLLI at Furman University (a private university) is highly integrated into the university. Furman University professors, as well as departments, routinely offer courses as a part of the OLLI program. They, along with many retired faculty and other experts in a broad range of areas, contribute to the overall amount of and quality of the OLLI. Special events are also coordinated/cosponsored by OLLI at Furman and other university units.This OLLI had the highest relocation percentage (46%) for OLLIs surveyed in 2016, which is reflective of the Greenville, SC community being attractive to men and women retiring from careers in other parts of the country. The OLLI program is an important source of new friendships as retirees move to the area and make the social interaction component of the program more important to participants than might be the case if they had lived in the area longer.	2,200 learners 500 members volunteer
OLLI at Northwestern University	OLLI at Northwestern University is a part of the School of Professional Studies, one of the 12 colleges that make up this private university. The OLLI has two program sites: Downtown Chicago and Evanston. The Chicago location is located on campus and the Evanston location is off campus, but members can participate at either site.Courses generally use the “discussion model” format, whereby peer members share responsibility for coordinating academic courses—planning/preparing course proposals, content research, and recruitment of a small group of discussion leaders for course sessions. This variation on delivery models leads to high value placed on discussion and discourse among the members over the more typical lecture style OLLI models. Content such as single-session lectures and special events are typically presented by active Northwestern faculty or community experts, and presented regularly but less frequently than discussion model courses.	1,400 learners, 940 Chicago location, 460 Evanston location, 320 members volunteer

Note. OLLI = Osher Lifelong Learning Institute.

These descriptions and membership statistics generate questions for future research on the value of lifelong learning. Future studies should explore variables such as institutional structure (public vs. private), levels of administrative support and advocacy, location on and off campus, and normative teaching approaches (discussion vs. lecture). Membership size and makeup are also likely influential and warrant further investigation. Finally, geographic and cultural variables should be explored in future research on the value of lifelong learning.

Limitations

This study concerned OLLIs, and future studies should look to other networks of lifelong learning providers. One reviewer of this article highlighted concerns regarding the weighted influence of the four highly respondent institutes (i.e., University of Delaware, University of Southern Maine, Furman University, and University of Kansas). A post hoc phi-correlation analysis revealed no significant or low correlational differences between how these four institutes had more or less codes/themes compared with the other eight institutes. The only notable significant difference is that these four institutes showed significantly fewer responses that contained multiple codes (Φ = −.181, p < .001). Overall, the study’s sample size was large, but homogenous. More heterogeneous samples will diversify this research on values.

Conclusion

Aside from the study findings, the study methods provide researchers and practitioners avenues to better understand the learners they serve. These methods can and should be replicated in surveys conducted both within and across lifelong learning institutes. Furthermore, the low correlations between each of the themes highlight their uniqueness as constructs, which should be investigated further in lifelong learning research. This study also provided insights regarding which aspects of lifelong learning and lifelong learning institutes are most important to older adults based on a national, cross-institute survey. Institutional structures and practices are tied to the lifelong learners’ values, but more research is needed in this regard in order to best serve and incorporate the burgeoning population of older adults across the world.

Footnotes

Appendix A

Appendix B

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Craig A. Talmage

Notes

Author Biographies

Craig A. Talmage is as a visiting assistant professor of Entrepreneurial Studies at Hobart and William Smith Colleges in Geneva, New York. He holds a PhD in Community Resources and Development from Arizona State University (ASU). He actively seeks to elucidate ways to understand and improve the well-being of all persons across the U.S. and world.

Robert Jack Hansen formerly worked as the Deputy Center Director for Research of the NASA Ames Research Center and as Associate Director for the Institute for Human and Machine Cognition. While he does work part-time with these organizations on occasion, he actively participates in the research of the National Resource Center for Osher Institutes and Osher Lifelong Learning Network. He holds both a BS degree from Stanford University and a ScD degree from MIT.

Richard C. Knopf is the Director of the Osher Lifelong Learning Institute at Arizona State University (OLLI at ASU), Director of the ASU Partnership for Community Development, and a professor of Community Resources and Development at ASU. He is actively engaged in the Age Friendly University movement. He seeks to advance quality of life in communities through innovative partnerships among multi-sector organizations. He has an MS in Urban and Regional Planning and a PhD in Environment and Sustainability, both from The University of Michigan—Ann Arbor.

Steven P. Thaxton serves as the Executive Director of the National Resource Center for Osher Institutes (Osher NRC) at Northwestern University. He holds a bachelor’s degree from the University of Minnesota, and had a 33-year career in television management. He joined the first OLLI at the University of Southern Maine (USM), from which he earned an MS in Adult and Higher Education. He actively manages the national research conducted by the Osher NRC.

Riley McTague graduated from Hobart College in Spring 2018 studying International Relations. He conducted this research as part of an Entrepreneurial Studies independent study project with Dr. Talmage in Fall 2017. He minored in Entrepreneurial Studies as well.

David Bennett Moore graduated from Hobart College in Spring 2018 studying Economics. He conducted this research as part of an Entrepreneurial Studies independent study project with Dr. Talmage in Fall 2017. He minored in Entrepreneurial Studies as well.

References

Armstrong

Gosling

Weinman

Marteau

(1997). The place of inter-rater reliability in qualitative research: An empirical study. Sociology, 31, 597-606. doi:10.1177/0038038597031003015

Boulton-Lewis

G. M.

(2010). Education and learning for the elderly: Why, how, what. Educational Gerontology, 36, 213-228. doi:1080/03601270903182877

Brady

E. M.

Cardale

Neidy

J. C.

(2013). The quest for community in Osher lifelong learning institutes. Educational Gerontology, 39, 627-639. doi:10.1080/03601277.2012.734147

Cain

T. K.

(2008). Elderhostel. In Loue

S. J.

Sajatovic

(Eds.), Encyclopedia of aging and public health (pp. 310-311). Boston, MA: Springer. Retrieved from https://link-springer-com-443.web.bisu.edu.cn/content/pdf/10.1007/978-0-387-33754-8_151.pdf

Chen

L. K.

Kim

Y. S.

Moon

Merriam

S. B.

(2008). A review and critique of the portrayal of older adult learners in adult education journals, 1980-2006. Adult Education Quarterly, 59, 3-21. doi:10.1177/0741713608325169

Clark

Fochs Heller

Rafman

Walker

(1997). Peer learning: A popular model for seniors education. Educational Gerontology, 23, 751-762. doi:10.1080/0360127970230801

Dauenhauer

Steitz

D. W.

Cochran

L. J.

(2016). Fostering a new model of multigenerational learning: Older adult perspectives, community partners, and higher education. Educational Gerontology, 42, 483-496. doi:10.1080/03601277.2016.1157419

Eyler

(2002). Reflection: Linking service and learning: Linking students and communities. Journal of Social Issues, 58, 517-534. doi:10.1111/1540-4560.00274

Formosa

(2014). Four decades of Universities of the Third Age: Past, present, future. Ageing & Society, 34, 42-66. doi:10.1017/S0144686X12000797

10.

Golding

B. G.

(2011). Social, local, and situated: Recent findings about the effectiveness of older men’s informal learning in community contexts. Adult Education Quarterly, 61, 103-120. doi:10.1177/0741713610380437

11.

Hansen

R. J.

Brady

E. M.

Thaxton

S. P.

(2016). Demographic and behavioral characteristics of Osher lifelong learning institute members. Journal of Continuing Higher Education, 64, 42-50. doi:10.1080/07377363.2016.1131541

12.

Hartmann

D. P.

(1977). Considerations in the choice of interobserver reliability estimates. Journal of Applied Behavior Analysis, 10(1), 103-116. doi:10.1901/jaba.1977.10-103

13.

Hebestreit

L. K.

(2009). An evaluation of the role of the University of the Third Age in the Provision of Lifelong Learning (Doctoral dissertation). Retrieved from http://uir.unisa.ac.za/bitstream/handle/10500/1498/thesis.pdf

14.

Horsburgh

(2003). Evaluation of qualitative research. Journal of Clinical Nursing, 12, 307-312. doi:10.1046/j.1365-2702.2003.00683.x

15.

Hsieh

H. F.

Shannon

S. E.

(2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15, 1277-1288. doi:10.1177/1049732305276687

16.

Illeris

(2003). Adult education as experienced by the learners. International Journal of Lifelong Education, 22, 13-23. doi:10.1080/02601370304827

17.

Jakobi

A. P.

(2012). International organisations and policy diffusion: The global norm of lifelong learning. Journal of International Relations and Development, 15, 31-64. doi:10.1057/jird.2010.20

18.

Jarvis

(2000). “Imprisoned in the global classroom”—revisited: Towards an ethical analysis of lifelong learning’. In Appleton

Macpherson

Orr

(Eds.), Lifelong Learning Conference: Selected papers from the inaugural international Lifelong Learning Conference (pp. 20-27). Yeppoon, Queensland, Australia: Lifelong Learning Conference Committee, Central Queensland University.

19.

Jenkins

(2011). Participation in learning and wellbeing among older adults. International Journal of Lifelong Education, 30, 403-420. doi:10.1080/02601370.2011.570876

20.

Katz

Earl

Ben Jaafar

Elgie

Foster

Halbert

Kaser

(2008). Learning networks of schools: The key enablers of successful knowledge communities. McGill Journal of Education/Revue des sciences de l’éducation de McGill, 43, 111-137. doi:10.7202/019578ar

21.

Kim

Merriam

S. B.

(2004). Motivations for learning among older adults in a learning in retirement institute. Educational Gerontology, 30, 441-455. doi:10.1080/03601270490445069

22.

Kirkpatrick

D. L.

(1994). Evaluating training programs: The four levels. San Francisco, CA: Berrett-Koehler.

23.

Lamb

Brady

E. M.

(2005). Participation in lifelong learning institutes: What turns members on? Educational Gerontology, 31, 207-224. doi:10.1080/03601270590900936

24.

Lee

J. M.

(2016). A comparative analysis of demographics and reported preferential learning modes of Florida and non-Florida Osher Lifelong Learning Institutes members (Doctoral dissertation). University of South Florida, Tampa, FL.

25.

Lietz

Zayas

L. E.

(2010). Evaluating qualitative research for social work practitioners. Advances in Social Work, 11, 188-202.

26.

Marsden

(2011). A study of the co-operative learning model used by the University of the Third Age in the United Kingdom. International Journal of Education and Ageing, 2(1), 55-66. Retrieved from http://eprints.lancs.ac.uk/51677/1/Marsden_IJEA_Vol_2_No_1.pdf

27.

McDonough

K. E.

Davitt

J. K.

(2011). It takes a village: Community practice, social work, and aging-in-place. Journal of Gerontological Social Work, 54, 528-541. doi:10.1080/01634372.2011.581744

28.

Merriam

S. B.

Kee

(2014). Promoting community wellbeing: The case for lifelong learning for older adults. Adult Education Quarterly, 64, 128-144. doi:10.1177/0741713613513633

29.

Oddi

L. F.

(1987). Perspectives on self-directed learning. Adult Education Quarterly, 38, 21-31. doi:10.1177/0001848187038001003

30.

Pstross

Talmage

C. A.

Peterson

C. B.

Knopf

R. C.

(2017). In search of transformative moments: Blending community building pursuits into lifelong learning experiences. Journal of Education Culture and Society, 62(2), 62-78.

31.

Pulakos

E. D.

(1984). A comparison of rater training programs: Error training and accuracy training. Journal of Applied Psychology, 69, 581-588. doi:10.1037/0021-9010.69.4.581

32.

Stemler

S. E.

(2004). A comparison of consensus, consistency, and measurement approaches to estimating interrater reliability. Practical Assessment, Research & Evaluation, 9(4), 1-19.

33.

Talmage

C. A.

Lacher

R. G.

Pstross

Knopf

R. C.

Burkhart

K. A.

(2015). Captivating lifelong learners in the third age: Lessons learned from a university-based institute. Adult Education Quarterly, 65, 232-249. doi:10.1177/0741713615577109

34.

Talmage

C. A.

Mark

Slowey

Knopf

R. C.

(2016). Age friendly universities and engagement with older adults: Moving from principles to practice. International Journal of Lifelong Education, 35, 537-554. doi:10.1080/02601370.2016.1224040

35.

Tornstam

(2011). Maturing into gerotranscendence. Journal of Transpersonal Psychology, 43, 166-180. Retrieved from https://thewisdomfactory.net/wp-content/Readings-for-show/Transcendence%20ageing.pdf

36.

Weinstein

L. B.

(2004). Lifelong learning benefits older adults. Activities, Adaptation & Aging, 28(4), 1-12. doi:10.1300/J016v28n04_01