Well-Being Predicts Digital Literacy Skills in Rural Older Adults

i AN ABSTRACT OF THE DISSERTATION FOR THE DEGREE DOCTOR OF PHILOSOPHY IN THE SCHOOL OF LIBRARY AND INFORMATION MANAGEMENT Brady D. Lund Presented on: April 6, 2022 Title: Well-being Predicts Digital Literacy Skills in Rural Older Adults Abstract approved: ____________________________________________________________ This study examined the relationship that personal, economic, and social-relational factors have with the development of digital literacy skills among rural older adults who have been impacted by the COVID-19 pandemic. Based on ecological theories proposed by Kim and Moen (wellbeing) and Williamson (information behavior) and using modified questions from existing, validated surveys of the target population (Health and Retirement Survey, Jones-Jang et al. study of digital literacy), this study surveyed older adults in rural, western Kansas. The findings of this study indicate strong interrelationships between personal, economic, and social-relational factors and the three digital literacy indicators (information literacy scores, trust in interpersonal information sources, and trust in mass media information sources) among rural, independently living older Kansans. As the rural, independent-living, older adult population is rarely studied in any discipline, let alone library and information science, this study also provides a unique contribution to the scholarly corpus of the field and may inform future research to examine the lives and information needs of rural older adults. ii WELL-BEING PREDICTS DIGITAL LITERACY SKILLS IN RURAL OLDER ADULTS By Brady D. Lund Emporia, Kansas 2022 --------- A Dissertation Presented to EMPORIA STATE UNIVERSITY --------- In Partial Fulfillment Of the Requirements for the Degree Doctor of Philosophy The School of Library and Information Management This research was partially funded by a generous grant from Emporia State University, The Boylan Scholar’s Award. iii ___________________________________ Wooseob Jeong, Ph.D. Dean of the School of Library and Information Management ____________________________________ Jerald Spotswood, Ph.D. Dean of the Graduate School and Distance Education ___________________________________ Mirah J. Dow, Ph.D. (Chair) ____________________________________ Brendan Fay, Ph.D. (Member) ____________________________________ Keith Wylie, Ph.D. (Member) iv Copyright © Brady D. Lund 2022 All rights reserved v ACKNOWLEDGEMENTS Thanks to my committee members and chair, Dr. Mirah Dow, for their work and support of this project. Their efforts are greatly appreciated. Thank you also to the support and oversight of Dean Wooseob Jeong, Dean Jerald Spotswood, and all of Emporia State University’s School of Library and Information Management and the Graduate School. Thanks to Emporia State University’s Boylan Scholar’s Award, and its benefactors, for a grant in support of my dissertation research, which enabled me to distribute my mail survey without incurring substantial financial hardship. I also want to thank my family, friends, and colleagues for their support and encouragement through my academic journey. Lastly, I want to extend my gratitude to all of the participants in this study and to the many intermediaries who assisted with the distribution of my survey. Without their willingness to participate, this study would not have been possible. vi TABLE OF CONTENTS Chapter 1: Introduction ................................................................................................................... 1 Background ................................................................................................................................. 1 Research Problem ........................................................................................................................ 4 Research Questions ..................................................................................................................... 4 Research Purpose ........................................................................................................................ 5 Conceptual Framework ............................................................................................................... 7 Significance of the Study ............................................................................................................ 8 Definitions of Terms ................................................................................................................... 8 Summary ................................................................................................................................... 11 Chapter 2: Literature Review ........................................................................................................ 13 Lifestyles and Well-Being of Older Kansans ............................................................................ 15 Rural Older Adults During the COVID-19 Pandemic .............................................................. 15 Studies of Information Behavior Among Older Adults ............................................................ 17 The Flow of Misinformation and Its Impact on Older Adults .................................................. 22 Use of Digital Technology Among Rural Older Adults ........................................................... 25 Digital Literacy and Rural Older Adults ................................................................................... 27 Theoretical Framework ............................................................................................................. 30 Application of the Framework in this Study ............................................................................. 38 Summary ................................................................................................................................... 39 Chapter 3: Methods ....................................................................................................................... 40 Research Paradigms .................................................................................................................. 40 Restatement of Purpose of the Study and Rationale ................................................................. 41 Research Questions and Hypotheses ......................................................................................... 42 Data Collection Instrument and Process ................................................................................... 43 Research Population and Setting ............................................................................................... 50 Data Analysis Procedure ........................................................................................................... 53 Reliability and Validity of Methods Utilized in this Study ....................................................... 57 Study Timeline .......................................................................................................................... 58 The Researcher .......................................................................................................................... 59 Ethical Considerations and IRB Approval ................................................................................ 60 vii Summary ................................................................................................................................... 60 Chapter 4: Results....................................................................................................................... 62 Descriptive Statistics ................................................................................................................. 62 Correlation Analysis .................................................................................................................. 65 Structural Regression Model ..................................................................................................... 67 Open-Ended Comments Analysis ............................................................................................. 69 Summary ................................................................................................................................... 74 Chapter 5: Discussion and Conclusion ..................................................................................... 76 Overall Interpretation of the Study’s Findings .......................................................................... 76 Central Research Question ........................................................................................................ 81 Theoretical Implications ............................................................................................................ 84 Further Evidence to Support Findings in the Present Study ..................................................... 85 Implications for Practice and Policy ......................................................................................... 89 Future Research ......................................................................................................................... 90 Study Limitations ...................................................................................................................... 92 Conclusion ................................................................................................................................. 94 References ..................................................................................................................................... 96 Tables .......................................................................................................................................... 120 Figures......................................................................................................................................... 132 Appendices .................................................................................................................................. 139 viii LIST OF TABLES Table 1…………………………………………………………………………………….……120 Table 2…………………………………………………………………………………….……121 Table 3…………………………………………………………………………………….……122 Table 4…………………………………………………………………………………….……123 Table 5…………………………………………………………………………………….……124 Table 6…………………………………………………………………………………….……125 Table 7…………………………………………………………………………………….……126 Table 8…………………………………………………………………………………….……130 Table 9…………………………………………………………………………………….……131 6 This study utilizes a cross-sectional survey approach, with both quantitative (multiple choice, Likert) items and qualitative (open response) items. The questions for this survey were adapted from existing, well-known and widely used surveys for the United States’ older adult population. In particular, questions adapted from the Health and Retirement Survey, a biennial study conducted by the University of Michigan’s RAND Center for the Study of Aging, were used to obtain demographic background information about participants (e.g., age, social activity, use of various digital communication technology). Questions from Jones-Jang et al.’s (2019) survey of media literacy and fake news susceptibility were used to examine participants’ level of digital literacy. The Jones-Jang survey included a scale that can be used to measure digital literacy skills. Permission was received from Jones-Jang on February 9, 2021 to use these scales in the present study. The survey instrument for this study was developed in SurveyMonkey, an electronic survey platform, and consisted of 50 questions arranged into four categories: personal resources questions, economic questions, social-relational questions, and digital literacy questions. The survey was distributed to older adults (age 60+) in western Kansas, which was defined, for the purposes of this study, as all counties within the state’s first congressional district. The survey was distributed in two ways: through mail, to a random sample of 300 older adults within western Kansas, and through the assistance of information intermediaries within the communities in this region, such as political figures, church leaders, and public librarians. Statistical analyses, including correlation analysis and structural regression modelling, were used to examine relationships among variables and compare these findings to those of the previous (HRS 2020 and Jones-Jang et al.) studies of a nationwide sample. 7 Conceptual Framework There are two major theories/models that inform the framework for this study: Williamson’s Ecological Theory of Information Behavior and Kim and Moen’s life-course, ecological model. In Williamson’s Ecological Theory of Information Behavior, the behavior of an information seeker is seen as the product of their personal factors as well as the information seeker’s interactions with their environment. The development of information behavior, like digital literacy skills, can be understood better, according to Williamson’s (1998) theory, by considering factors including lifestyle, social and cultural values, socio-economic circumstances, work situation, personal/biological characteristics, affective/spiritual influences, and physical environment Kim and Moen’s (2002) life-course, ecological model is the second influence for this study’s framework. Kim and Moen’s model suggests that the behaviors of older adults are the product of personal resources (i.e., innate characteristics), economic resources, and socialrelational resources. In the Kim and Moen model, all factors (divided amongst the three categories of resources) may influence one another as well as the behaviors of individuals. The personal resources category consists of age, gender, subjective ratings of health, sense of life control, morale and depression. The economic resources category consists of employment status (i.e., retired, part-time employed, full-time employed), income, and finances (i.e., accumulated wealth and spending). The social-relational resources category consists of level of social engagement, quality of relationships, and technology access and media consumption. These factors are all anticipated to influence older adults’ behaviors including information behavior, the intentional or unintentional glimpsing, encountering, or avoiding information (Case & Givens, 2016). As such, the Kim and Moen (2002) model, paired with Williamson’s conception of information behavior, directly influences the research questions, hypotheses, survey questions, 8 and analysis for this study, in that the study examines the role of these factors in influencing digital literacy skills, including whether and to what extent a correlation exists among these factors, is examined. Significance of the Study This study is significant in that it directly addresses the needs and behaviors of a population that is often overlooked in research and social service interventions: the rural older adult who lives independently (not in an assisted living environment). As noted in this introduction, the rural older adult population has experienced significant barriers to finding and using reliable information during the COVID-19 pandemic, beyond the challenges they already faced with the proliferation of misand disinformation. Perceptions of quality-of-life and wellbeing have declined. Yet, this population and its information behaviors during the pandemic have been all but ignored. By understanding factors involved in rural older adults self-identified acclimation to the changing digital environment, it may be possible to develop more-targeted digital literacy initiatives for rural older adults, which will ultimately improve the well-being of both the individual and the rural community as a whole. Definitions of Terms The following are definitions and/or descriptions of terms and concepts central in this study including well-being, misinformation, and digital technology and literacy. These terms are outlined here and are integrated throughout the dissertation. Older Adult For this study, an “older adult” is anyone age 60 or older (60+) (Kansas Department of Aging, 2018). While it may be argued that using age to define social groups is problematic— after all, one may act much younger than their age and may be physically healthier than the 9 average person of a certain age—it is nonetheless an important way to classify members of a society. Even though individuals live much longer and healthier lives today than in past generations, there are fundamental differences between an individual whose chronological age is 60 years versus someone whose age is 40 years. In order to ensure that the experiences of all these groups of individuals are investigated, it is important to distinguish among them as is done in this study. Well-Being In this study, a definition of well-being provided by Kim and Moen (2002) is utilized. This definition posits that well-being is a state of being content with one’s life. Kim and Moen (2002) propose that well-being is the product of three categories of social-emotional capital: personal resources, economic resources, and social-relational resources. Each of these three categories of resources are defined below. Personal Resources. A measure, according to Kim and Moen (2002), of mental, physical, and emotional well-being, comprised of the following factors: age, gender, health (subjective perceptions of both physical and mental health relative to the “average”), and sense of life control (control over what happens in one’s life, sense of autonomy). Economic Resources. A measure, according to Kim and Moen (2002), of financial wellbeing, comprised of the following factors: employment (whether one is employed full-time, parttime, retired, or unemployed), income (poverty status/lower, middle, upper class income), and finances (accumulated wealth/value of owned property). Social-Relational Resources. A measure, according to Kim and Moen (2002), of social and relationship well-being, comprised of the following factors: social engagement (how active one is within their community and activities), the quantity of one’s relationships (size of family, 10 number of friends), and the quality of one’s relationships (how strongly one feels connected to their family and friends). Misinformation and Disinformation Though there are many unique definitions of misinformation, each with subtle differences, the definition used for the purposes of this study is based on Stahl (2006, p. 86), where misinformation is simply defined as “wrong or misleading information.” As noted by Cooke (2017, p. 14), this misinformation can emerge from a variety of sources and take a variety of forms, which often makes it difficult to identify. Misinformation is related, and often considered synonymous, to “disinformation,” with the latter term generally being used to specifically denote information that is known to be wrong but is still deliberately spread (Cooke, 2017). Just as all information contributes to individuals’ knowledge and beliefs, misinformation contributes to false knowledge and misguided beliefs. These outcomes can lead to harmful consequences to individuals and the whole of society. Digital technology Digital technology is defined as any technology that transmits data in a digital format (i.e., “zeroes and ones”) (Salmons & Wilson, 2009). This includes any technology that connects to a network, including the Internet. Examples of digital technologies that may be relevant to the population in this study (rural older adults) include personal computers, cell phones, ‘smart devices’ (watches, speakers, and Internet of Things technologies like voice activated services), and virtual reality, as well as specific services on these technologies, like email, web browsing, and social media. Digital literacy 11 The American Library Association’s Digital Literacy Task Force (2020) defines digital literacy as, “the ability to use information and communication technologies to find, evaluate, create and communicate information, requiring both cognitive and technical skills.” Children are often the focus of digital literacy research (Sefton-Green & Erstad, 2009); this generation (“Gen Z”) is considered “digital natives,” or individuals who have grown up in the digital age and always used digital technologies. Older adults, conversely, are not digital natives and have often been overlooked in discussion of importance of digital literacy concepts. Summary This study consisted of a cross-sectional survey approach to learning from rural Kansas older adults about their access and use of information during the COVID-19 pandemic. To address validity and reliability of this study, the survey questions were adapted from standardized, nationwide surveys. It was distributed in western Kansas (in the state’s first congressional district) via mail, and with the assistance of community intermediaries, trusted individuals who can share and vouch for the quality and importance of the survey. After an approximately three-month period of data collection, the data was analyzed as appropriate to address the research questions for the study. Chapter Two of this dissertation is a review of literature that is necessary to fully address major concepts relevant to this investigation including the information behaviors of older adults (particularly those residing independently within rural and semi-rural communities), the unique barriers to finding reliable information faced by this population, and digital literacy and susceptibility to misinformation, and emphasizes both the history and pronounced gaps in the research surrounding older adults’ use of digital technology to find and use information. Chapter Three presents the methodological assumptions and approach for the study, including how the 12 data will be analyzed from both quantitative (statistical analysis) and qualitative (content analysis) perspectives. Chapter four presents the findings of this study and evaluates the research hypothesis. Chapter five discusses the significance of the findings, their relevance to existing literature, and provides concluding comments. 13 Chapter 2: Literature Review Though information behavior has long been considered an important area of study within library and information science, literature communicating research-based evidence in this area remains an opportunity for further research. For example, studies relevant to the information experiences of older adults—particularly those in independent-living situations—is sparse. Many of the studies of this population that do exist are merely descriptive (e.g., “these are the three most common information needs”) rather than explanatory (e.g., “this is why these are the three most common information needs”). These studies of older adults offer very little if any context or environment to human information behavior; instead, they too often describe the individual as though they exist in a vacuum. This type of approach is problematic and occludes the diversity of experiences and perspectives of individuals including those of older groups. This study focuses specifically on the rural older adult in the midwestern United States, a group that has been little studied and differs significantly in regard to their information environment and social relationships compared to the urban older adult. Even those urban older adults located in the same region (e.g., Wichita, Kansas) have shown differences in behavior from rural (e.g., Wilson, Kansas) peers. As noted by Norris-Baker and Scheidt (2005), the rural older adult is found to derive much of their social identity from their community. Rural older adults often see themselves not as “from” a certain community but rather as “part” of it. Older adults have often been described as the single-greatest sharers of “fake news” online (Rampersad & Althiyabi, 2020). The popular consensus has held that this population is simply more easily fooled and lacks critical thinking abilities due to cognitive decline. However, recent research has suggested a substantial social role in the drive to share misinformation 14 (Scherer & Pennycook, 2020). Fake news may be shared as a way to ignite a discussion or debate among friends and family in order to achieve some sense of social connection. Further, Brashier and Schacter (2020) indicate that older adults—who are not digital natives—lack knowledge of digital literacy skills but are not incapable of learning them, as social interventions by the researchers were able to successfully increase these skills in a sample of older adults. This finding supports the notion that older adults’ social relations could play a role in reducing their susceptibility to misinformation. During the COVID-19 pandemic, many older adults were left isolated as they sheltered at home to avoid the highly contagious coronavirus. This event very likely altered these individuals’ information access. Those who lacked access to digital technology, or had limited digital literacy skills, suffered greatest from intermittent states of information poverty and an influx of fake news, culminating in what Xie et al. (2020b) termed as an “information crisis” (p. 1419), where the lack of reliable information results in other negative consequences like poor health decision-making. These consequences demonstrate the social imperative that is supporting the digital literacy skills of the older adult population. This review of the literature emphasizes both the history and pronounced gaps in the research surrounding older adults’ use of digital technology to find and use information. It contextualizes the challenges that this population faces when acquiring and using platforms like Facebook to share and receive information. It also proposes a theoretical framework, structured around Williamson’s (1998) ecological theory of information behavior and Kim and Moen’s (2002) ecological model of post-retirement behavior, which may be used to construct an understanding of the socio-psychological phenomena that lead to a lack of digital literacy among a rural, older adult population that has been affected by the COVID-19 pandemic. 15 Lifestyles and Well-Being of Older Kansans In Kansas, about one-third of older adults—defined as individuals age 60+—reside in a rural community (United States Census, 2019). Another one-third live in semirural environments (areas that have characteristics of rural and urban communities—generally population of around 50,000 or less). Unlike most urban dwellers, older adults in rural communities often derive a major part of their social identity from the community in which they live (Wright et al., 2016). This makes members of this population feel more attached to the people around them and a more traditional way of living and less likely to be “early adopters” of new ideas and technologies (Hollifield & Donnermeyer, 2003). They value the relationships they have with other community members and place a high level of trust in their judgments. Internet access and literacy is more limited (Hargittai et al., 2019; Hodge et al., 2017), which limits access to quality rural healthcare information and social contact (Cotton et al., 2013; Currie et al., 2015). This population has also been regularly shown to be among the most susceptible to believing in and sharing misinformation on the Internet (Dodson et al., 2015; Brashier & Schacter, 2020). These attributes of rural older adults influence or comprise their “social identity,” the qualities, beliefs, and personality of individuals (Jenkins, 2014). Rural Older Adults During the COVID-19 Pandemic Relevant information intended for rural older adults is not always disseminated through the channels most apt to reach them, as they have been found to have different information source preferences as compared to younger adults and children for both vital, health-related needs as well as everyday information (Gilly & Zeithaml, 1985; Capella & Greco, 1987; Su & Conaway, 1995; Eriksson-Backa, 2008; Williamson & Asla, 2009; Hallows, 2013). Simply put, reliable information that older adults need is not reaching them (McIntosh et al., 1995; Pettigrew, 22 The Flow of Misinformation and Its Impact on Older Adults The harm of misinformation on the lives of rural older adults is felt not only in these individuals’ inability to participate as informed members of society but also on the impact that misinformation has been shown to have on overall health and well-being of this population. From an economic perspective, research dating back over half a century has indicated detrimental financial impacts due to a lack of information. Nelson (1970) and Rao and Bergen (1992) indicate that a lack of information about economic conditions and price comparisons result in spending more money on products than is financially prudent or sustainable. A lack of information literacy—the ability to find, evaluate, and use information to address a need (regardless of format, as opposed to digital literacy)—related to nutritional matters, as discussed by Spiteri-Cornish and Morales (2015), results in poor health and nutrition behaviors. Additionally, even when information is plentiful, misinformation can cause significant issues. Misleading information in reviews and product descriptions may lead older adults to make poor financial decisions (Dou et al., 2012; Malbon, 2013; Zhu & Zhang, 2010). Misinformation that causes fear or mistrust in the economic and/or political system can cause significant harm to the decision-making capacity of individual (Jerit & Barabas, 2006). Similarly, a lack of reliable information can have major impacts on health, particularly during a pandemic like COVID-19. Lack of information has led people to make decisions that result in serious health maladies, like cancer and poor nutrition (Autier et al., 1994; Roberts, 2013). Additionally, misleading health information has been shown to be valued by consumers— in part because it capitalizes either on fears of the consumer or promises incredible outcomes (Syed-Abdul et al., 2013; Gastil & Marriott, 2019). This enticing information can come with potentially deadly consequences. Several studies have looked at misinformation about vaccines in light of the 2008-2009 H1N1 pandemic and the 2020 COVID-19 pandemic. A commonality 23 among these studies is that a lack of access to reliable information, combined with misinformation that propagates fear and uncertainty, leads to the rejection of potentially lifesaving medical advice and services—testing and vaccinations (Ahmed et al., 2011; Zheng et al., 2020; Keelan & Kumanan, 2007; Wen et al., 2020). Belief in misinformation about medical topics can lead to a distrust of credentialed medical professionals (e.g., the director of the National Institutes of Health) and misplaced faith in the sources of misinformation or figures that trumpet it (Hatcher, 2019; Murray et al., 2003). In a time of considerable political divisiveness in the United States, research has also shown that a lack of information or confluence of misinformation, combined with limited critical thinking/political reasoning, can increase susceptibility to fake news and the likelihood of supporting political positions that are irrational or harmful to the individual. Lupia (1994) suggests that Americans, in general, fail to seek information needed to form complete opinions on political topics for which they are voting, while Cruz et al. (2021) recently found similar insights among their sample voters in the Philippines. A lack of an informed voter base was shown by Bartels (1996) to influence outcomes of elections; specifically, incumbents performed an average of five-points better in elections where voters were considered to be poorly informed about the candidates, while Democratic candidates performed an average of two points better in these circumstances. Xie and Jaeger (2008) note that older adults appear to be disinterested in political participation online, preferring in-person debate and discussion. Pennycook and Rand (2019) found that susceptibility to political fake news is more related to non-critical thinking and lack of conflicting sources of information than particular political beliefs, suggesting that isolation and a reduction in information source could increase susceptibility to political misinformation. 24 Lack of information can also result in social withdrawal and isolation (Hand et al., 2017). Withdrawal and isolation have consistently been shown to have tremendous effects on both physical and mental health, including mortality among older adults. Steptoe, Shankar, Demakakos, and Wardle (2013) found that social isolation is correlated with an increased risk of cardiovascular disease, cognitive deterioration, and mortality, regardless of whether the isolated individual actually feels lonely or depressed by the state of isolation. Holt-Lunstad et al. (2015) found that the increased likelihood of early mortality among socially isolated individuals is between 26-32% depending on factors such as whether or not the individual feels lonely in their isolation. Hawton et al. (2011) found that individuals who were social isolated rated worse on both subjective and objective measures of health and well-being. Misinformation is usually not different from any other type of information in terms of form (which makes it challenging to the average person to identify) but is often spread very differently (Del Vicario et al., 2016). The Internet is ripe ground for the sharing of this type of information. Anyone can publish content on the Internet, with no standards for the veracity of information shared, and social media has enabled this information to be shared easily and virtually instantaneously (Berghel, 2017). This makes misinformation much easier to produce and disseminate than true information. Generally, the content of misinformation is also disputatious, making it more enticing to share as a point of conversation (not unlike gossip). Apuke and Omar (2021) noted that sharers of misinformation about the COVID-19 pandemic were motivated by a variety of factors, including “altruism” (a genuine belief that what they share will be helpful to others), information sharing, information seeking, socialization (simply for the purpose of socially engaging with others), and to pass time. Because there is a diversity of motivations that support the sharing of fake news, and many are related to social 25 engagement rather than an informational role, it is difficult to make any practical effort to mitigate its spread. These factors all contribute to a rapid spread of misinformation that makes it one of the most dangerous and pressing challenges in the world today. These outcomes all illustrate the malicious impact of misinformation on one of society’s most vulnerable populations. This research, however, is incomplete. There is a lack of knowledge about how rural older adults during the COVID-19 pandemic, in particular, were forced to cope with a deluge of misinformation that threatened their health and safety. This study will examine potential correlates of misinformation susceptibility among the rural older adult population to determine to what extent this information may have caused serious problems in other areas of these individuals’ lives. Use of Digital Technology Among Rural Older Adults Psychological and information science research has, in recent decades, taken an interest in how older adults adopt and use digital technology to acquire information, an approach that has refined and, arguably, added rigor to the study of older adults’ information behavior. Neil Charness and his colleagues, primarily at Florida State University, have examined the relationship between social factors in aging and the use of digital technology (Charness et al., 1992; Charness et al., 2016; Charness & Boot, 2009; Charness & Holley, 2004; Czaja et al., 2006; Czaja et al., 2018). This research began in the 1980s and has continued to this day, having a substantial impact on subsequent work in this area, in that it has identified the role of social support and attitudinal factors in determining whether older adults adopt technology. An experimental study performed by Chevalier et al. (2015) examined the capacity of younger and older adults to conduct successful web searches using a search engine. Not only did older adults struggle to find the needed information for difficult questions, but they also 26 exhibited fewer strategies to obtain information and had less confidence in their abilities to execute successful searches. The searching behavior/queries demonstrated a limited knowledge of everyday information retrieval practices that for many younger adults are common knowledge and taken for granted. Consequently, older adults were more likely to seek help from human mediators than their younger adult counterparts. However, given that many younger adults are digital natives (having grown up using digital technology), they were undoubtedly socialized into using these digital technologies and employed more advanced information retrieval strategies, while older adults may not have been. The Internet is an important facilitator for the exchange of interpersonal information, through social media websites like Facebook. Sinclair and Grieve (2017) found that Facebook usage was a significant source of social connectedness for older adults, on par with the level of connectedness experienced by younger individuals. This social connectedness, in turn, helps improve psychological outcomes for these older adults. Chen and Schulz (2016) found that use of information and communication technologies (like the Internet and social media) were correlated with diminished social isolation by improving connections to the outside world, growing social support, engaging individuals in activities of interest, and boosting selfconfidence/self-efficacy. Use of information and communication technologies, however, is skewed towards individuals with higher education and socioeconomic status. As noted by Hargittai and Dobransky (2017), in addition to Palsdottir (2012a; 2012b) and Xie, Watkins, Golbeck, and Huang et al. (2013), older adults with higher education and income are generally more privileged in terms of online information access and use behavior, with significantly higher levels of Internet use skills and information retrieval practices. 27 While recent developments in Internet technology have helped to diminish the digital divide to an extent, there is still a significant gap in the amount, relevance, and quality of information produced and available for older adults compared to other populations residing in the United States (Williamson, 1998; Padilla-Gongora et al., 2017). As noted historically by researchers from a wide variety of disciplines, older adults tend to adopt new technologies later than those in other age groups, or remain “laggards,” using the diffusion terminology of Everett Rogers (Gilly & Zeithaml, 1985; Rogers, 1986; Selwyn, 2004; Vicente & Lopez, 2006; Ferro et al., 2010; Gonzalez et al., 2012; Caceres & Chaparro, 2019). This challenge is exacerbated by the fact that, historically, very little information has been created that is relevant to the information needs and uses of older adults (Hales-Mabry, 1993; Baker, 2004). Research findings indicate that this has left this population reliant on information sources that may be more restrictive (such as close family and friends, or television stations) and less reliable. This is not a problem caused by the older adult, but rather a failure on the part of communities and information infrastructures to support this population with their unique needs. Unlike children and younger adults, who were “socialized” into using digital technology essentially from birth, older adults were born in a different time when none of this technology existed and were occupied by work, family, and other obligations that consumed time when these technologies were introduced. Digital Literacy and Rural Older Adults Older adults have repeatedly been shown to be the greatest source of fake news susceptibility and sharing (Rampersad & Althiyabi, 2020). Traditional studies of misinformation susceptibility and older adults have used brain imaging to examine areas of brain activation during information searching tasks. Research by Roediger and Geraci (2007) points to a direct 28 relationship between aging and susceptibility towards misinformation, due to changes in brain anatomy. The ability to evaluate information occurs through a complex communication interplay between the neurons in the basal ganglia and the frontal cortex, which is responsible for the brain’s executive functioning (autonomous decision-making ability). As noted in Gottlieb et al.’s (2014) study, and further substantiated in Roediger and Geraci’s (2007) study, the brain experiences normal atrophy with aging that impacts both the basal ganglia and the frontal cortex. This may diminish the brain’s ability to discriminate between reliable and unreliable sources of information. Roediger and Geraci’s study found that individuals with particularly low frontal lobe functioning were significantly less likely to identify misinformation than individuals with higher frontal lobe functioning status, suggesting that the extent of atrophy, through normal brain aging and/or dementia, reduces the information literacy abilities of the individual. Several recent studies have pointed to deficits in digital literacy skills as being a key factor in why older adults, on average, are more likely to believe and share misinformation. Scherer and Pennycook (2020) note that, while popular opinion might argue that older adults are more susceptible to misinformation due to cognitive decline, it is, in fact, a lack of knowledge about digital information and how well news can be manipulated or outright fabricated, that leads to this susceptibility. As with the adoption of digital technology (discussed in the prior section) older adults have not been socialized into understanding the risks and threats associated with online information, while many younger adults have had this benefit through their schooling and occupational training. While changing information behaviors among older adults may be, in part, due to neurological change, it is also hypothesized to be partly (perhaps even largely) due to the older adult’s changing role in society (i.e., from employed to retired; from active, daily participant, to 29 more isolated and desirous of using technology to maintain social connections). Older adults have been found to use social media for different purposes than younger adults, in ways that may make them more susceptible to sharing misinformation and lacking digital literacy skills. Brashier and Schacter (2020) note that older adults, unlike younger generations, use technology to connect with others (their community/family) rather than to gather new information. They show greater concern for younger generations and can see sharing news as a way to connect and help inform—not necessarily realizing the harm they are doing by sharing false information. They may share a provocative story simply to initiate a conversation. This population is more trusting, particularly when it comes to information provided by people they know (Lund & Long, 2021; Pennycook & Rand, 2020). If they deem someone trustworthy, they are less likely to ever question whether something they say is a lie (Castle et al., 2012). These recent perspectives on why older adults may be more susceptible to misinformation are important, as they indicate that there may be social reasons, as opposed to only cognitive decline-based conclusions for this susceptibility. Unlike cognitive decline, social behaviors can be altered or reversed with education and training (Birdsong & Freitas, 2012). However, they are complex behaviors to understand and identify how to best address. The COVID-19 pandemic has, in many ways, proven to be a litmus test for digital literacy skills Those who lack these skills are much more likely to eschew preventative measures like mask wearing and express skepticism about vaccines, which, of course, also puts them at higher risk of contracting the disease, suffering from severe symptoms, and spreading the disease to others (M Chen et al., 2020; Y Chen et al., 2020; Wang, 2020). Studies that have emerged during the pandemic have begun to provide clues about the types of individuals who are most susceptible. One of the strongest of these factors is age. As discussed previously, biological 30 changes due to aging may play some role in information behavior, while major life transitions like retirement also play a substantial role. The older an individual, the less likely they are to have developed sufficiently strong digital literacy skills to negotiate the modern ecosystem of mis-/disinformation overload (Chu et al., 2021; El Skarpa & Garoufallou, 2021; Lund & Ma, 2021; Zhao et al., 2020). Within the older adult population, there are many additional factors that have been shown to predict digital literacy skills. Religiosity and political ideology are two such factors, with the very religious and politically conservatives expressing greater COVID skepticism (Choma et al., 2020; Leibovitz et al., 2021; Miller, 2020; Motta, 2021; Taylor et al., 2020). Married men were found to be more susceptible to misinformation in the studies of Al-Mohaithef and Padhi (2020), M Chen et al. (2020), and Ng (2020). Lazarus et al. (2020) and Taylor et al. (2020) identified low educational attainment as a correlate of low digital literacy skills. Romer and Jamieson (2021) found that the sources and types of information consumed were important, with those watching mainstream television news having the best ability to parse through COVID misinformation. Theoretical Framework The framework for this study is based on two ecological theories: one from gerontology and one from information science. The first of these is Williamson’s Ecological theory of information behavior. In this theory, the behavior of an information seeker is seen as reflective of some personal/innate factors (e.g., age) as well as the information seeker’s interactions with their environment. The development of information behavior, like digital literacy skills, can be understood better, according to Williamson’s (1998) theory, by considering factors including 31 lifestyle, social and cultural values, socio-economic circumstances, work situation, personal/biological characteristics, affective/spiritual influences, and physical environment The second is Kim and Moen’s (2002) life-course, ecological model. Kim and Moen’s model suggests that the individual is the product of both innate characteristics (age, gender, intellect) and environmental forces (both social relationships/bonds and external factors like mass media and physical environments). Based on these factors, the individual develops an identity that directly relates to their behaviors (including information behaviors). The framework outlines a collection of features both innate and environmental that can be examined empirically in relation to a specific information behavior, such as digital literacy. Origins of Ecological Theory Bronfenbrenner’s (1979) ecological theory has historically been used to describe the role of a child’s environment on their development; however, in recent years, elements of the theory have been adapted to describe the development of behaviors in a diversity of populations. According to the theory, there are several “systems,” or layers of influence that impact the development of human behavior. Individuals most frequently and directly interact with their microsystem, which is comprised of the individuals and communities that individuals elect or are forced to interact with on a near daily basis. For older adults, this might include their spouse (and in some cases, only their spouse), church, senior center, friends that they meet each morning at the coffee shop, etc. Beyond the microsystem is the mesosystem, comprised of individuals with whom the individuals in the microsystem interact (i.e., friends of friends, family of friends, coworkers of family). The individuals in the mesosystem exist in the microsystem of members of your microsystem, but not in your microsystem. 38 They inform and support the proposed relationship in this study between well-being (based on Kim and Moen’s model) and digital literacy. Application of the Framework in this Study For the purposes of this study, Williamson’s theory and Kim and Moen’s model together serve as a clear framework of how personal and ecological influences can impact human behavior. While Kim and Moen’s model was initially developed to describe general postretirement behavior of older adults, it fits well with the development of the specific behavior of digital literacy, as discussed in the information behavior research of Williamson and others. This framework invites several testable propositions about the development of digital literacy skills that inform the research hypotheses for this study including that • individuals derive their identity, which includes a set of normative information behaviors and value judgements of information needed and information quality, from the convergence of personal attributes (e.g., gender) and environmental influences (social interaction). • the identity that individuals subsume guides the acquisition of skills and behaviors, including digital literacy skills. Personal, economic, and social-relational factors may all influence whether digital literacy skills are successfully acquired by rural older adults. • these personal, economic, and social-relational factors and their relationship to digital literacy skills can be evaluated quantitatively, using correlation and SRM analyses, and qualitatively, by examining how individuals describe their information searching patterns. This is based on the approach that Kim and Moen both use in their own works to examine older adults’ behaviors. 39 This framework and the above propositions, in addition to informing the research questions, guide the selection of the survey methodology for the study and the questions incorporated in said survey. Each section of questions within the survey is based on one of four elements, digital literacy abilities and the three mechanisms of well-being (personal resources, economic resources, and social-relational resources). Each sub-question under each of the three mechanisms aligns with one of the specific resources (e.g., employment status, sense of life control, quality of relationships). This ensures that there are no superfluous or unnecessary questions. Each question is connected to a specific aspect of the theory being evaluated, as discussed in the methods chapter that follows. Summary The rural older adult population has often been overlooked in information behavior studies (and across the entire landscape of scholarly research). This population is one of the most disproportionally afflicted by a lack of access to reliable information. They take pride and derive identity from their small communities, which are slowly diminishing and dying during this period of urbanization. Solutions to the challenges this population faces are sparse, even among the agencies tasked with supporting them. This study seeks to uncover the relationship between personal, economic, and social-relational factors and the digital literacy abilities of rural older adults. 40 Chapter 3: Methods This study employs a mixed methods design utilizing a cross-sectional survey research strategy. The survey contains both quantitative (multiple choice and Likert) items and qualitative (open-ended response) items. In total, 50 questions were included that span four main topical areas related to the research questions: those related to personal resources, those related to economic resources, those related to social-relational resources, and those related to digital literacy. Quantitative data will be analyzed in SPSS using common statistical analyses of significance (Correlation, Regression/SRM). Open-ended response data will be analyzed following a content analysis procedure, which will allow for the identification and quantification of themes within the participants’ responses. The research approach is informed by a pragmaticconstructivist epistemology. Specific details of the survey are informed by the theoretical framework, discussed in the prior sections as an ecological-based understanding of the development of information behaviors. Research Paradigms In any research, it is important for the researcher to state clearly their philosophical assumptions and personal perspectives and biases related to the research. As the researcher in this study, I subscribe, epistemologically, to the pragmatic constructivist framework (Talja et al., 2005). As discussed below, this framework rejects both realist and anti-realist notions of reality, suggesting that a reality exists beyond human experience that can be examined objectively, but that this reality is nebulous and constantly changing. For this reason, the pragmatic constructivist epistemology emphasizes research driven by the specific research questions under investigation and the use of a variety of methods (and the mixed-method approach in general) to coordinate and triangulate findings. As for the researcher’s role and perspectives, the researcher intends to 41 take a neutral stance towards the subject matter being studied, though acknowledges several personal experiences that color his perspective on the subject of the research as discussed in the following subsections. These assumptions drive the selection of the research questions and methods selected for this study as well as the data analysis and interpretation techniques utilized. Research Methodology For this study, a cross-sectional survey approach, wherein data are collected at only one point in time, was utilized. This survey incorporates both quantitative and qualitative elements. The following subsections discuss the nature and assumptions of qualitative and quantitative studies and relevant procedures and risks that were considered while developing the questions and structure of this survey. Combining Qualitative and Quantitative Research Elements in a Survey Mixed methods research attempts to bridge the strengths and weaknesses of both quantitative and qualitative methods (Brannen, 2016). Mixed methods allows the researcher to address research problems/questions at different levels, to both describe and provide explanation for phenomena. All research approaches have some innate bias; a mixed methods approach reduces the extent to which bias may skew or invalidate findings based on the methods selected alone. While the survey in this study collects primarily quantitative data, several open response questions are included to allow greater detail for respondents in describing their information experiences. Restatement of Purpose of the Study and Rationale The purpose of this study is to examine the extent to which characteristics of rural older adults living in independent settings (age, educational attainment, socioeconomic status, technology availability, and social connections) relate to the development of abilities to find and 42 use information in electronic formats, also known as digital literacy skills (American Library Association Digital Literacy Taskforce, 2018). Social and psychological factors, which, according to Kim and Moen (2002), include (among others) age, health, employment status, and quality of interpersonal relationships, are proposed as contributors and/or predictors of the acquisition of digital literacy skills. This study evaluated these factors, relative to rural, independently living older adults, to determine the validity of the Kim and Moen (2002) theoretical model by quantitatively measuring the relative strength of each factor as they contribute to the development of digital literacy skills. Research Questions and Hypotheses Based on the gaps identified in the literature, and stated in the research problems, this study is guided by one central research question, which, in turn, guides the development of the survey instrument and data analysis procedures: • To what extent does a sense of well-being predict digital literacy skills among older adults living in rural Kansas during the COVID-19 pandemic? o To what extent do economic factors help predict digital literacy skills? o To what extent do personal factors help predict digital literacy skills? o To what extent do social-relational factors help predict digital literacy skills? Based on the theoretical framework and research problem and questions identified for this study, the following hypothesis and sub-hypotheses have been developed that provide insight into its anticipated outcomes: 1. Well-being will be shown to have a strong statistical relationship with the digital literacy skills of individuals: 43 • H1 0 = Well-being has no relationship with the digital literacy skills of rural older adults. • H1 a = Well-being has a statistically significant relationship with the digital literacy skills of rural older adults. Sub1. Economic resources of rural older adults will be shown to have a strong correlation with the digital literacy skills of individuals: • H2 0 = Improved economic resources have no relationship to the development of digital literacy skills. • H2 a = Improved economic resources relate to increased digital literacy skills. Sub2. Personal resources of rural older adults will be shown to have a strong correlation with the digital literacy skills of individuals: • H3 0 = Improved personal resources have no relationship to the development of digital literacy skills. • H3 a = Improved personal resources relate to increased digital literacy skills Sub3. Social-relational resources of rural older adults will be shown to have a strong correlation with the digital literacy skills of individuals: • H4 0 = Improved social-relational resources have no relationship to the development of digital literacy skills. • H4 a = Improved social-relational resources relate to increased digital literacy skills Data Collection Instrument and Process Impact of COVID-19 Pandemic on Selection of Data Collection Approach 44 Initially, the researcher had intended to utilize a mixed-methods approach, which would have occurred in two phases: a survey of rural older adults and then interviews with a smaller subset of these respondents. The COVID-19 pandemic made the second phase of this study difficult to execute. Due to the nature of this study’s subjects and the vulnerability and safety risks to the older adult population, both in-person interviews and technology-facilitated/Zoom interviews were not feasible in the opinion of the researcher’s dissertation committee, though they believed that the importance of the topic warranted investigation even despite this change in methods. For these reasons, it was determined that the best approach for this study would be to focus on the survey aspect, while future research post-pandemic may enhance the qualitative aspect by including interviews as part of the data collection. Selection of the Survey as Data Collection Instrument The survey method, and specifically the questionnaire approach, was selected as the data collection method for this study, as it is best suited to collect both the large amount of quantitative data needed for statistical analysis and the qualitative data that can add context and depth to the study’s findings. The survey method is one of the oldest and well-established research methods, having its birth in the population censuses and election polling in the 18 th and 19 th centuries (Rossi et al., 1983). Among the benefits of the survey method are that they are easy to distribute among a large population, allow greater flexibility to participants (they can decide when and where to complete it), and all participants answer the same, structured set of questions. Among the few limitations of the method are the limited ability to clarify/explain the meaning of questions as well as survey fatigue (people being asked too often to complete surveys). These weaknesses are mitigated by including questions that have been successfully 45 used in past studies and by using a population that is not accustomed to receiving many survey participation requests. Data Collection Procedures The survey in this study was distributed over a period of roughly three months. Two methods of participant recruitment were used. The first of these distribution methods was via mail. 300 copies of the survey were printed and mailed by the researcher, along with an introductory letter and a copy of the informed consent statement. A copy of the introductory letter included with the mail survey is provided as Appendix 1. Funding support for the mail survey’s distribution was provided by Emporia State University’s Boylan Scholars’ Award. For the mail survey, 300 individuals from the first congressional district (discussed in study population section below) were selected using information provided on Whitepages.com. The selection of survey recipients was made with the assistance of a random number generator. From the 62 counties in the first congressional district, 300 were randomly chosen (i.e., of the 300, Lyon County may have been randomly selected five times, while Gove County was randomly selected only three times). A list of names was created using a similar procedure with a list of the 100 most common first and last names in the United States (Social Security Administration, 2021; United States Census Bureau, 2021a). The names and counties were then randomly sorted (so “John,” “Smith,” and “Lyon County” may get randomly sorted together). A visual depiction of this sorting process is shown in Figure 2. Finally, an individual with that name living in that county, who was age 60+, was selected to receive the mail survey. If no one with that name lived in that county, then a new name was selected until a suitable recipient was found. 46 Second, “Intermediaries,” community members in rural towns who are well-connected and trusted, were used as a point-of contact in dissemination of the survey to local participants. Intermediaries include county extension office employees, church leaders, city council members, and local business leaders. A list of the intermediaries selected for this study is provided as Appendix 2. This approach to participant recruitment was suggested by Rick Scheidt (personal communication, October 13, 2020), a professor of environmental gerontology and human sciences at Kansas State University, who has conducted several extensive research studies involving the older adult population in western Kansas over a period of 50 years. These intermediaries were identified from websites of churches, local governments, and county extension offices. They were provided information about the study and the researcher, a link to the survey in electronic format, and the opportunity to ask any questions they have about the project. They were asked to share the survey (in electronic or print format) to individuals in their community whom they believe would be interested in participating. The questions for the survey instrument were adopted from two existing, nationwide surveys, both of which have been validated through repeated use and are respected among the scholarly community. The first of these surveys is the Health and Retirement Survey (HRS). This survey has been conducted biannually (every other year) since the 1980s by the University of Michigan’s RAND Center for the Study of Aging. Each version of the HRS receives about 20,000 responses from a representative sample of the United States’ older adult population. The questions included in the survey vary from year-to-year but—given ongoing world events with the COVID-19 pandemic and political and health misinformation—in 2020 took on an emphasis on pandemic response and digital technology. Several questions focus on family and community relations, including the sense of closeness that older adults feel with these groups. These 47 questions tie directly to the research questions for the present study and offer a source of validity as to the quality of the question wordings and triangulation of findings. The raw data from the HRS is available freely online and is often used for secondary data analysis in gerontological research studies. For the present study, this data set was used as a source of comparison between the nationwide dispositions (HRS data) and dispositions of a specific, western Kansas population (the present study). A small subset of questions relating to digital literacy were adapted from the second survey. This survey was Jones-Jang et al.’s (2019) survey of media literacy skills and susceptibility to misinformation/ “fake news.” This survey included a series of questions designed to assess the digital literacy skills of participants, which are used for a similar purpose in the present study. These questions offered a validated assessment of digital literacy (as opposed to using questions solely developed by the researcher) and the data produced from them will serve as a point of statistical comparison between the findings for a general population (Jones-Jang et al. study) and a specific, western Kansas older adult population (the present study). The resulting survey instrument for this study was created within SurveyMonkey, an electronic survey platform, but was distributed both in print and electronic format in order to remove Internet access as a barrier to participation. A full copy of this resulting survey instrument is included as Appendix 3. Many of the questions for the survey came directly from the Health and Retirement Survey and the Jones-Jang et al. survey and are used for comparative purposes; these questions are identified below within parentheses (e.g., “(HRS)”). An informed consent statement is included at the beginning of the survey that includes details about the study and contact information for the researcher (see Appendix 4). Participants must indicate that they 54 (dependent variables). The analyses for addressing this sub-question will isolate the relationship between responses to these questions, using the correlation matrix as well as discussing in greater detail findings from the SRM specific to the influence of economic resources on a sense of well-being and the presence of digital literacy skills. For sub-question 2, pertaining to the relationship between personal resources and digital literacy, analysis will focus on responses to questions 1, 2, 8-12, 19, and 20, relating to age, gender, health, and life control, and the digital literacy questions (30-46). The analyses for addressing this sub-question will isolate the relationship between responses to these questions, using the correlation matrix as well as discussing in greater detail findings from the SRM specific to the influence of personal resources on a sense of well-being and the presence of digital literacy skills. For sub-question 3, pertaining to the relationship between social-relational resources and digital literacy, analysis will focus on responses to questions 3, 5, 13-18, and 22-29, relating to sociability and relationships, and the digital literacy questions (30-46). The analyses for addressing this sub-question will isolate the relationship between responses to these questions, using the correlation matrix as well as discussing in greater detail findings from the SRM specific to the influence of social-relational resources on a sense of well-being and the presence of digital literacy skills. Normality, Data Distribution, and Statistical Tests Utilized in this Study Following data collection, the researcher performed normality tests on the data. ShapiroWilk tests were performed for each dependent variable. The results of these analyses indicated that the data exhibits a slight skew (is not normally distributed), with an average S-W statistic across all variables of .954 (p < .01). Accordingly, non-parametric statistical analyses are employed in this study. The Kruskal-Wallis H Test, the non-parametric equivalent of the 55 Analysis of Variance (ANOVA) test, is used to analyze differences in means. The Spearman Correlation test (non-parametric alternative to the Pearson correlation) is used to analyze relationships among variables. The Ordinary Least Squares Regression is used for regression analyses. Structural Regression Modelling Utilized for the purpose of evaluating the theoretical model for this study is structural regression modelling. SRM is used by researchers to evaluate predictive research hypotheses, such as the claim that wealth and education can predict incarceration rate (Kelloway, 1995). Wealth and education are latent variables, in the sense that they cannot be easily examined based on a single survey question and instead are best evaluated as a construct comprised of several related questions (e.g., wealth might include questions about annual salary, total savings, total debts). SRM is an extension of path analysis, a type of regression modelling that allows variables to be compared bidirectionally (e.g., impact of salary change on wealth and wealth change on salary), with more than one dependent variable compared at a time (Streiner, 2006). This allows for the interrelationships among all variables to be modelled in a reader-friendly format (see Figure 5). The initial SRM model is a hypothesis to be tested—it displays the relationships as suggested by the theoretical model used for this study. The final model (after the data analysis has been performed) will include regression coefficients that indicate the relative influence of each variable on all other variables. SRM is beneficial for testing theory, as opposed to regular correlation or regression analysis, which does not clearly indicate the direction of relationships among variables and/or does not allow for computing relationships among latent variables. SRM begins with a theory, which is based on existing literature. A model to be tested is proposed based on this theory. A 56 series of regression analyses are then performed to test the proposed model. SRM is particularly common in information systems and technology research, where researchers want to infer relationships among latent variables, variables that are not directly measured or observed (Lund, 2021). Figure 5 displays the factors (and their interrelationships) that are examined using SRM. This proposed model was developed by the researcher based on Kim and Moen’s theory. Note that many of the factors collected through individual survey items (e.g., “physical health” and “mental health”) combine to form a single construct (e.g., “health”). These constructs, as discussed above, are formed using confirmatory factor analysis. These constructs then form together to demonstrate relationships to specific traits (e.g., digital literacy). The relationship (correlation) between items on each layer (factor, the constructs, and the trait of digital literacy skills) are calculated and displayed along the lines connecting these items. This form of modelling both makes clear how all survey questions align with the testing of the theoretical model and offers a clear display of the findings (whether and to what extent the relationships among these variables exist). The research hypotheses reflected in this model would be that all relationships (as represented by the arrows) are positive. For most ratio and ordinal-type variables, correlation analyses were also performed to identify the relationships more clearly between each variable/sub-question of the survey. Content Analysis Procedures Open-ended response data (questions 47-50) was transferred to NVivo for analysis. A content analysis approach adapted from Marshall and Rossman (2006) was used to analyze the data, with the assistance of the NVivo platform: • Both team members read all the survey responses to get a sense of the responses. 57 • Both team members re-read the data marking all direct responses to the question. • Looking for patterns in the data, direct responses are organized into categories that emerge. • Each response is sorted into the appropriate category using the respondents’ actual words. • The results are reviewed, looking for overlap and redundancy and to refine and revise the category titles. • From the survey, instances of verbatim narrative were selected to illustrate categories. A fellow PhD candidate fulfilled the responsibility of team member for this analysis. Following the initial coding of all data, measures of intercoder reliability were calculated. The team members had over 98% agreement on their initial codes and an interclass correlation coefficient of .978 (p < .01). The high level of agreement is likely due to the unambiguous distinctions among the codes selected for the study. In cases were disagreement occurred among the two coders, the team members discussed until they could agree on a single code. As noted in a prior section, a benefit of qualitative data is that it captures perceived reality as described in the participants own words. Experienced qualitative researchers, with familiarity with the topic of this dissertation, are helpful in eliciting meaningful findings from the raw qualitative responses. This data will help to substantiate and add depth to the quantitative findings. Reliability and Validity of Methods Utilized in this Study The reliability and validity of the methods utilized in this study are supported by the adoption of questions from existing, popular surveys of the target population (older adults). Both 58 the Health and Retirement Survey and the Jones-Jang et al. survey have been shown to produce reliable data without leading to confusion or fatigue among the older adult respondent population. Having existing data sets from these previous studies further allows for the comparison of results. If responses to the survey of rural older Kansans differs substantially from those of a nationwide population, then this will clearly indicate to the researcher that further exploration of the data is necessary to determine whether some mistake occurred in the collection of data, or if there is simply a statistically significant difference in the responses of a rural older Kansan compared to the “average” older adult. Either way, the researcher is not completely “left out in the woods” to determine whether the survey questions truly examine what the researcher intended. Statistical analyses of quantitative data remove much of the researcher’s interpretation biases that might otherwise occur with the descriptive presentation of data. The interpretation of the qualitative data following established content analysis standards supports an analysis that presents a less biased perspective of the findings’ significance, relative to an analysis that does not utilize interpretation guidelines. While both survey method and both quantitative and qualitative data collection have clear limitations, they are considered by researchers to have a strong level of validity and reliability for attaining meaningful and accurate findings (Williams & Webb, 1994; Appleton, 1995). Study Timeline The project took approximately four months to complete from the initiation of data collection to the completion of the data analysis. The breakdown of each phase of the project’s execution is as follows: 59 1. The administration of the survey through mail and intermediary formats. The mail surveys were printed, packaged, and mailed on May 14, 2021. For intermediaries, an initial email was disseminated on May 31, 2021, with a follow-up email on June 14, 2021. The email included a request to share a link to the survey with anyone they believe would be interested in completing it. There was also a flyer attached to the email that could be printed off and displayed to advertise for the survey. The survey was available for participants to complete for a period of approximately two months. All mail and electronic survey responses were due by August 1, 2021, so that analysis of the data could commence. 2. Following the collection of the data (concluding on August 1), analysis was performed using SPSS and NVivo. The researcher was well-versed in these platforms and, as such, the learning curve/time needed to use these platforms was minimal. The analysis was completed over the course of approximately two weeks (completed August 13, 2021). 3. After data was analyzed, approximately five months were be needed to complete the writing of the results, discussion, and conclusion sections of the dissertation. This phase also allowed time for reading and input from dissertation committee members (completed March 1, 2022). The Researcher The researcher’s perspective towards this research topic, population, and setting is informed by his own experience spent living with his grandparents for several years during his early post-graduate education. Through this experience, he experienced first-hand how older adults found, collected, and shared information amongst one another as well as the role that the community played in the life of the rural older adult. This research is driven by the goal of identifying key areas of concern in the information experiences of rural older Kansans and 60 proposing solutions to promote information equity. However, the approach to this study is decidedly objective. While the topic and approach are informed by personal experiences of the researcher, all collection and analysis of the data is performed—as much as possible—with the researcher in the role of neutral examiner. This approach is driven by the preference of the researcher, which is demonstrated in the types of research questions and methods selected for the study. Ethical Considerations and IRB Approval An application for approval to use human subjects was submitted by the researcher on April 6, 2021. The requested time period for data collection was from May 1, 2021 through April 30, 2022. This IRB application was approved on April 22, 2021. A revised IRB application was submitted on April 30, 2021, which reflected the additional mail survey methodology (made possible through a research grant received by the researcher) and was granted approval on the same day. A copy of the IRB approval letter is included as Appendix 7. Summary This study consists of a survey approach to examining the role of personal and social/ecological factors in the development of digital literacy among rural older adults in western Kansas. This survey incorporates both quantitative and qualitative items. It was distributed over the course of three months across the first congressional district on the state of Kansas (stretching from Emporia/Lyon County in the east, to the Kansas-Colorado border in the west, and Hutchinson/Reno County in the south). In addition to mailing surveys to a randomly selected group of older adults in the first district, intermediaries in the rural communities were asked to help disseminate information about the survey and encourage individuals to participate. The data collected from the survey would be analyzed using SPSS (quantitative data) and NVivo 61 (open-ended response data). Statistical and content analyses allow the researcher to evaluate the hypotheses proposed based on the study’s theoretical framework. 62 Chapter 4: Results The purpose of this mixed-methods study is to better understand how factors in rural older adults’ lives relate to their acquisition of digital literacy abilities and usage of digital technology during the COVID-19 pandemic. The study provided measures of the extent to which a sense of well-being among rural older Kansans predicts their digital literacy skills. Additionally, the relationship between the constructs of personal, economic, and social-relational resources with the information literacy skills and digital technology adoption of rural older Kansans during this unprecedented period is explored. Over the summer of 2021, a 50-item survey, informed by the work of the Health and Retirement Survey (2020) and Jones-Jang et al. (2019), was used to question participants about their information experiences during the time of the COVID-19 pandemic. Invitations to participate and consent forms were mailed to 300 possible participants in Kansas’s First Congressional District. Additionally, thirty intermediaries were randomly selected to serve as local community leaders who could help to distribute the survey in print and electronic format. Valid survey responses were received from 206 participants. Measures of central tendency, correlation analyses, and structural regression analyses were conducted for each of the study’s quantitative variables. Qualitative data was analyzed using standard content analysis procedures. This chapter presents the results of each of these analyses. Descriptive Statistics Table 3 shows the distribution of respondents for three of the scales used in this study (with “sense of life control” and “social connectedness” being calculated as the average of a respondent’s answers to a set of five questions). With two of the scales—sense of life control and finances—a left-skewed distribution is present (most respondents indicated a very high level of 63 life control and a better than average financial situation). However, with the other two scales— social connectedness and health—near-perfect normal distributions are present, where equal numbers of respondents fall on both sides of the scale (with the mid-point of 3). Interestingly, while distributions were similar for respondents under age 70 and over age 70 for two of the scales, they were quite different on the sense of life control and finances scales. On the “sense of life control” scale, a large percentage of respondents over the age 70 indicated a very strong sense of control (62%) than those under the age of 70 (43%). One possible explanation for this difference is employment status, as those over the age of 70 were much more likely to be retired, which may give a greater sense of control over one’s life compared to having to work a 40 hour a week job. However, those in the under age 70 group were more likely to indicate that are able to make ends meet “very well” (28% compared to 8%), likely due to preversus post-retirement income and financial demands. Shown in Table 4 are the frequencies with which respondents participate in a variety of social (involving active communication/interaction with other people) and nonsocial (can be done alone) activities. Respondents participate in nonsocial activities with greater frequency than social ones, with most participating in these nonsocial activities daily versus weekly or monthly for social ones. The older the respondents the less likely they were to indicate that they use social media or browse the Internet on a regular basis, and the more likely they were to indicate that they regularly read books/magazines/do puzzles daily. Younger respondents who were still employed full-time were less likely than their peers who were retired to indicate that they regularly participate in group activities, but younger respondents in general were more likely to indicate that they participate in group activities than the older respondents. 70 information, such as asking “around town” or asking family and friends. This was particularly true for information related to local community information (e.g., a new business moving into town). The difference in the percentage of responses involving use of the Internet to find information is statistically significant, based on a chi-square test, X 2 = 20.5, p < .01. For the question relating to the price of a new computer, 59% of responses related to some use of a computer; for the operating hours of a grocery store, 56% of responses involved the Internet; and for the new store example, 45% of responses related to the Internet. Significant differences in these types of information across the three questions may explain the differences in sources used by older adults. For the first question, about finding information about a computer, it is worth noting that a computer is a product, not a local service or event. It can be readily (and, often, most conveniently) be purchased online. Many of the responses to this question discuss online search strategy, such as respondent 36, who said, “Type computer costs in the research bar and it will pull up various types of companies with cost of computers. I can go to the sites and check out more information.” Some participants used the Internet along with interpersonal sources, such as respondent 50, who said, “I check store sites and line up the capabilities I need before choosing what I will purchase. I check with siblings who are in this business as well.” The most common responses were that they simply “Google it,” but a few respondents mentioned specific online retailers they use, such as, “Costco” (respondents 3, 116), “visit Dell.com for best prices” (respondent 7), “Staples.com” (respondent 20), “Amazon.com” (respondents 20, 41, 43, 67, 80, 126, 130, 135, 175, 179), “Honey to find deals” (respondent 30, 106), and “Woot” (respondent 116, 125). Among those who would visit a 71 brick-and-mortar store, most listed one of two specific retailors: Walmart (5 responses) or Best Buy (9 responses). Similarly, though the grocery store in the example was a local store, even local stores’ operating hours are generally posted somewhere online (whether it is Facebook, the yellow pages, or a full-fledged website), and it is simply more convenient to search Google for the operating hours than drive down to the store or call someone. “Google it” was again the most common response to this question. However, a significant number of respondents indicated that there was no case in which this information would be necessary, because they know the hours. As respondent 4 said, “I only have two grocery stores (in town) to choose from. They have had the same opening hours for as long as I have lived here,” respondent 31 stated that they had, “Been going there for over 60 years, I think I know,” while respondent 162 noted, “I just know what time they open! Or, if there's a question, I'll call and listen to the recording on their business hours.” A few respondents noted that this was not a relevant information need for them because, as respondent 87 said, “I know my grocery stores are all open by the time I wake up.” The development of a new business moving into town, however, is local. There is unlikely to be much posted online unless the business is a major chain. The word “heard” may also be key. Information about this development is already circulating around town, so sources within town may well have more information than any sources online. This question received the most answers in which the respondent provided multiple search strategies. For instance, respondent 63 noted that, “I would check with the local chamber of commerce or do search the website of the business,” while respondent 99 said that “I would ask family or friends if they heard about it, but also search online to see if I could find information.” This suggests that the respondents may anticipate more barriers or difficulty in finding information to address this 72 need. However, there were still a fair number of respondents that indicated that they would use the Internet, though fewer indicated that they would use Google, instead suggesting they would use the “local newspaper or store’s website” (respondent 169) or their “local city’s Facebook page” (respondent 189). In alignment with the research questions that guided this study, there are several notable differences in responses to this study based on the demographics of respondents. Those who were older named the Internet as an information source less frequently and named fewer information sources in general (less likely to provide any response or only provide a single source rather than multiple). Socially connected individuals were more likely to list multiple sources of information for these questions (e.g., ask around town and use the Internet). More financially stable individuals appear to shop more using online sites like Amazon and are more likely to check with business and city sources on new developments within the city. Additional Comments Made by Participants Participants were given the opportunity, with the final question, to supply any additional comments they have related to the study. A total of 59 substantive comments were recorded. The most common of these comments was along the lines of “I can use my computer with no difficulty” or “I use my computer to find information all the time” (21 comments). Four commenters said that they had been using the Internet for a long time (“decades”) and struggled to remember a time before computers. One commenter noted that they regularly use a computer for work and their computer has become vital to their daily functioning. The second-most common type of comment (seven instances) was that the participant prefers traditional computers to mobile devices, like respondent 105, who said, “I hate cell phones, I miss my easy land-line.” Several of these commenters (who tended to be in the older 73 group of participants) noted that the smaller screens on mobile devices made them difficult to read and/or navigate. Additionally, two commenters noted that they prefer print resources over digital resources and another two commented that they never, or very rarely, use computers. Three commenters said that they feel intimidated by computers, such as with the fear of breaking something or being clueless about where to even start with their search. As noted by respondent 75, “It's a little scary on what you can find out and do on computers and smartphones! I feel intimidated by them at times and then when they don't work. I just want to toss them out the window! I strongly feel that society relies too much on technology.” Another four respondents said that, depending on the type of information sought, it could be very frustrating to use computers at times, such as respondent 184, who said, “Usually it’s frustrating to find information online. There’s too much to look through and lots of distractions built into the system.” Offering a solution to some of the technology issues they face, four commenters suggested that more technology courses offered specifically for older adults would be helpful to reduce barriers. For instance, respondent 13 suggested that, “Classes regarding smartphone usage would be beneficial. I tried reading the manual but there were so many acronyms I didn't know that weren't defined that it became fruitless.” Respondent 93 similarly noted that, “It would be nice to have someone with knowledge to come to the seniors in town to teach them some simple things about technology.” In terms of information access to/on the Internet, in addition to the frustrations discussed above, four commenters said that broadband upgrades were badly needed in their community, as summed up by respondent 33, “Rural communities (most) do not have broadband which limits accessibility to internet access. Data is used to access the internet from smartphones. Rural 74 communities need broadband to stay in tune with current events.” Another commenter said that a recent upgrade to the broadband infrastructure in their community significantly improved their Internet access. One commenter stated that their only way to access the Internet was through using the computers at their local public library (though they prefer it that way, because “they can help me!” with computer skills). Three commenters noted that they have vital health information needs that they did not feel they could satisfy due to a lack of access to the Internet and to medical professionals, such as respondent 51, who stated that they “have many existing medical problems,” but, “struggle getting any long-term care insurance information.” Lastly, a few commenters added thoughts about social media as an information source. One noted that social media is their primary information source. Another said that they do not use social media and believe it is harmful. Another two commenters said that they were skeptical of information on social media, but that this was the most convenient information source to be found. One of these two commenters noted that they always read through the comments on a social media post and use this to make up their mind on a topic. Summary This study has identified significant relationships between factors of well-being and digital literacy, the ability to use information and communication technologies to find, evaluate, create and communicate information. Additionally, significant relationships were found between digital literacy and personal, economic, and social-relational resources. These findings provide support to reject the null hypothesis that well-being and digital literacy are not related. In fact, the results offer evidence to support that digital literacy should be considered a major component of well-being in the modern, digital society—even within the more-isolated lives of rural older 75 adults. The following discussion section will discuss the implications of these findings and situate them within the existing literature. 76 Chapter 5: Discussion and Conclusion This study has revealed many findings regarding the relationships between aging, demographic factors, and digital technology use and ability among older rural adults. The structural model for the study shows a strong predictive ability between personal, economic, and social-relational factors (well-being) and digital literacy skills, the ability to use information and communication technologies to find, evaluate, create and communicate information. Only a few variables (i.e., gender, participation in nonsocial activities) were found to play no significant role in the model hypothesized initially in the theoretical framework section. This bodes well for the extension of Kim and Moen’s model which, as this research has shown, incorporates for the digital literacy skills of rural older adults and has implications for future policy and practices for reaching older adults living in rural communities with vital information. Overall Interpretation of the Study’s Findings This study reveals several novel findings about the information behavior of a rarely studied population: rural, independent-living older adults. In regard to information behaviors, this study found that there are different levels of trust in interpersonal and mass media information sources, whereby interpersonal sources are generally better trusted by older adults than mass media sources; as individuals age, lower information literacy scores are found, with individuals in the age 70+ range having significantly lower scores than those under the age of 70. Personal, economic, and social-relational resources were all found to influence the digital literacy skills of rural older adults, while personal resources significantly influence perceptions about society and information sources, economic resources influence technology exposure and use, and social-relational resources influence trust in various sources of information. This section 77 provides details about each of these findings and their implications within the context of this study. Different Levels of Trust in Interpersonal and Mass Media Information Sources This study reveals important findings about the information sources trusted by rural older adults during the time of the COVID-19 pandemic, which has implications for social services, political, public policy, and media production. During the pandemic, in particular, these findings may help inform an approach to better reaching rural communities with vital information about the pandemic and busting misinformation about vaccines. It is widely known that many individuals residing in rural communities have largely avoided the guidance of health agencies like the Centers for Disease Control and Prevention (CDC) and have been some of the most resistant to vaccination (Alcendor, 2021; Kricorian et al., 2021). It is evident from the results of this study—which align with those of Williams, Ames, & Lawson (2019)—that most sources of authority enjoy low levels of trust among older adults residing in rural communities. Even print media, long an important source of news and information in rural communities, is treated with substantial skepticism. Many people, in fact, place greater trust in strangers than they do the politicians elected to serve them. Television news has near-neutral favorability, but likely only because of a note to respondents that this refers to their favorite news station (in fact, even though they were not asked, many respondents made it clear by handwriting on the survey “Fox News” or “only OANN”). Given the widespread disdain towards Dr. Anthony Fauci—the chief medical advisor to the President of the United States—and the CDC, the differential of +29 for experts/scientists may be surprising. However, it is important to note that several survey respondents indicated that these individuals are not seen as fitting solely in the “experts/scientists” category. Rather, these 78 figures were seen as being political, due to their role in the government. For instance, Dr. Fauci’s role as advisor to the President makes him seem like a political figure, even though his responsibility is to his scholarly knowledge of medicine and the ethics of his profession. So, while experts/scientists are generally trusted, the political appointment of an expert may interfere with the public’s trust of the appointee. In fact, perhaps the most trusted figure—outside of one’s friends and family—is more likely to be a well-educated medical professional, someone who may be known to the public but is not known for controversial political or medical stances— someone that is seen as being a highly-relatable member of the community. In other words, Dr. Fauci could be viewed as a disseminator, while a member of the community may be utilized as a diffuser of information to this population. Lower Information Literacy Scores with Aging Scores on the information literacy scales reveal that most participants in this study have limited information literacy skills. The majority scored “average” (a score of 3 on a five-point Likert scale), while the overwhelming majority of the remaining participants scored “below average” (2 or 1). It was further revealed that age has a significant relationship with information literacy scores, with significantly more individuals in the age 71+ group scoring a 2 or 1 on the information literacy scale, which ranges from 1 to 5. Those who were employed at the time of the study also scored better on the information literacy scale, perhaps due to greater social interaction, exposure to more diverse ideas, and use of computers for work-related activities. These deficits in information literacy skills among older adults represent a major concern in a world that is increasingly emphasizing the role of digital technologies and is increasingly plagued by the rapid spread of misinformation. Understanding the nature of how information is 79 produced (typically, in a biased way) and the role of disseminators in shaping the information that users consume will only become more important in years to come. Educational programs for rural older adults may be helpful. However, as noted in the prior section, the public standing of the individual who is presenting the program, and the level of trust they enjoy within the community, is key. Sources of educational programming that are perceived as politically biased will likely be disregarded. Recruiting trusted community members (perhaps including rural librarians themselves) could be useful in developing a program that members of the rural community will actually attend. Specialized training, that focuses on aspects of digital literacy that are identified as relevant in this study (particularly, focusing on improving information literacy scores) is requisite. General technology training, that lacks any focus on these key elements of information literacy, is insufficient. Personal Resources Significantly Impact Digital Literacy and Perceptions About Society Age appears to be an important factor for many of the other study variables—sense of life control, information literacy scale scores, trust in interpersonal sources—but not all. Participation in nonsocial activities, like reading news and browsing the Internet, is more frequent among older respondents, and this fact is likely reflected in the greater trust in mass media sources of information. Interpersonal sources were less trusted, perhaps due to a growing sense that people are “out to get you”—e.g., family planning to send an elderly relative to an assisted living facility—or the belief that family is belittling their beliefs due to their age. Similar findings to both of these phenomena have been regularly reported within existing literature (Donohue et al., 2009; Gordon, 2020; Klemmack & Roff, 1984; Lindquist et al., 2018). 86 Predictors of the Use of Digital Technology Among Older Adults In Lund (2022), several of the same variables that were considered in the present study were evaluated using correlation and regression analysis with data from the nationwide Health and Retirement Survey (HRS). In fact, Figure 5, presented earlier in this study as an example of structural regression modelling, is based on data retrieved from the Health and Retirement Survey. In this section, similarities and differences in findings for relevant variables in the nationwide Health and Retirement Survey and this survey of rural older Kansas is provided. Not all variables/questions align for both studies, so only relevant variables are included in this discussion. Nonetheless, several compelling relationships in the data are uncovered. Both the analysis of nationwide HRS data and the data for the present study of rural older Kansans indicate significant correlations between financial status and social activity and health. In the analysis of HRS data, socioeconomic status was found to have strong, positive correlations with self-esteem (.39), physical health (.41), sociability (.18), and life control (.41). In the present study of rural older Kansans, “finances” was similarly found to have positive correlations with health (.41), participation in social activities (.22), and life control (.14). Both studies also found that wealth did not correlate to the adoption of digital technologies by older adults. These findings demonstrate the similarities between rural Kansans and general older Americans in terms of the impact and relationships of financial status and well-being, thus lending further support to the revised models presented. There is, however, one significant difference in the finances-impact findings for the study. In the HRS study, no relationship was found between socioeconomic status and “relationship closeness” (.04 correlation), but in the rural older Kansans study, finances was found to have a significant correlation with connectedness (.38). Perhaps this difference indicates an outsized role of wealth in sense of belonging in rural versus urban communities. In other 87 words, in rural communities, wealth appears to be related to how connected one feels with their community (those who are wealthier feel more strongly that they are an important part of the community), while no such relationship exists across the general population. Alternatively, a factor that may be at play is that rural communities have less economic stratification (there are fewer “wealthy” people who live in rural communities) than urban communities (Goetz et al., 2018). The Lund (2022) study, while being exploratory rather than theory-driven, does provide some additional support to another core aspect of the present study’s theoretical base. In the HRS study, three factors were found to contribute directly to the use of digital technology for communication and all were highly correlated with one another: intellectual curiosity (openness to ideas/information from a variety of sources), self-esteem, and sociability. Sociability as a construct contained many of the same items/variables as social-relational resources in the present study, while self-esteem was essentially an amalgam of personal (life satisfaction, life control) and economic (socioeconomic status and work demands) resources. These components together are very similar to the four categories of resources that are hypothesized, and later confirmed statistically, to influence an overall sense of well-being. The overall similarities in these two studies of different populations of older adults provide support of greater generalizability of the final proposed model. Information Practices and COVID Beliefs Among Older Adults In the Lund and Long (2021) study, it was found that five factors contribute to increased skepticism of COVID and vaccine information: relationship status (similar to the “family relationships” variable in the present study), family influence on information behavior, gender (examined in the present study, but found to have a nonsignificant impact), family closeness 88 (similar to the “connectedness” variable in the present study), and political leaning (not examined in the present study, but mentioned in some of the qualitative comments). That socialrelational factors were found impactful both in the present study of digital literacy and the Lund and Long (2021) study of information gaps supports its validity as an important factor in the presence or absence of digital literacy skills of older adults. “Family influence on information behavior” has some relationship to “trust in interpersonal sources of information” in the present study, but the former is more direct. It asked participants “does your family influence what information you consume?” rather than “do you rely on family for information?” Again, this provides further support to the role of family in digital literacy skills of older adults. A unique contribution of the Lund and Long (2021) study is that it examined political leanings, finding that those who identified as politically conservative were more skeptical of the danger of COVID-19. While political leanings were not examined in present study, we know from recent election results that rural older Kansans tend to be very conservative (McKee, Ostrander, & Hood, 2017). Conservative political learnings could be yet another factor that influences the low information literacy scores of the respondents to the present study’s survey. These findings all add further evidence to support the above-proposed model of digital literacy and well-being. Information Behavior of Rural Older Adults During COVID-19 In the Lund and Ma (2021) study, interview participants were asked to describe the process of finding information to satisfy needs that arose during the COVID-19 pandemic. The findings indicate that, while many participants certainly gathered information to satisfy personal needs, there was also a social motivation behind information seeking. Many of the participants in the study noted that they tried to find answers to other peoples’ questions in order to give them a 89 reason to engage socially. There was a definitive relation between sociability and information behavior during this period of time. This, again, provides further support to the findings of the present survey study. Additionally, the Lund and Ma study examined from where in participants’ environments information was found. While Williamson’s past studies have emphasized the role of wider social networks for seeking information, Lund and Ma (2021) found that the pandemic restricted access to these individuals, requiring older adults to turn more frequently to close family and mass media. In the present study, it was found that digital literacy scores suffered among older adults, but those who scored the highest had the greatest trust in interpersonal and mass media sources of information. Given that the COVID-19 pandemic restricted access to fellow human beings, and that social life plays a major role in digital literacy scores, it can be predicted that many older adults suffered from their poor digital literacy skills during the pandemic. Indeed, qualitative findings from the present study, as well as the results of the Lund and Ma (2021) study, show that many (but certainly not all) struggled more than usual in finding the information they needed during the pandemic. Implications for Practice and Policy Given the findings in the present study for trusted sources of information among rural older Kansans, there may be some implications for educating the public about major health crises like the COVID-19 pandemic. While individuals with political appointments are at-risk of distrust by some people in society, health professionals and experts, according to this study, are still generally highly regarded. For example, it is likely possible that a “Ask a Local Medical Expert Your Questions About COVID-19” session at the public library, where a local physician or county health officials could be invited to answer questions about the virus and vaccines, 90 would be effective. Within the library, it may also be advisable for librarians to use language that avoids potentially politically loaded phrasings like “the CDC recommends…,” as also indicated as best practice by Gostin (2018). If done carefully (knowing the community that is being served), it may be possible to lend some credibility to these educational efforts. Librarians could instruct public library users in using reliable information provided by county health agencies to inform their knowledge about the virus. Several of the findings of this study also suggest that renewed investment in rural communities may improve technology adoption and digital literacy. The investment of more money into rural communities to improve living conditions and create more community engagement may improve both economic and social-relational resources, increase well-being, and perhaps create a fertile environment for encouraging older adults to acquire new digital literacy skills. Many rural older adults seem to be using digital technology with frequency, but it is likely through interaction with others that they gain digital literacy skills. Future Research There are many opportunities for future research based on the findings of this study. Here, I will focus on four avenues for further study that can dovetail nicely from the findings of the present study. Relating to digital literacy and well-being of rural older adults, one question that may be investigated is whether the findings for rural Kansas (a Midwestern state) hold for other areas of the United States and around the world. One might hypothesize that rural older adults in coastal states (more politically liberal states with perhaps more social support services available) could have different patterns of behavior, as could individuals in less technologically advanced and wealthy nations; these studies would allow researchers to confirm whether the coefficients identified in the present study hold true for other populations. 91 This study illustrates that a specific information behavior—digital literacy—can be better understood through the environmental and behavioral contexts in which it occurs. This study shows that, not only is this approach to studying information behavior feasible, but it also greatly enhances our understanding of how these complex behaviors manifest. While a traditional study may have looked at the relationships on the right side of the structural model (between information literacy scores and trust in information sources)—and this would still be a compelling study—of course it is all the more compelling and relevant to the work of interdisciplinary researchers to connect these behaviors to the construct of well-being. Further research may employ this broader contextual approach to further examine information behavior among older adults as well as other populations. Two related questions that are not addressed in this study are how digital literacy is acquired and how best to teach it to rural older adults. This study’s findings indicate what type of person is more likely to have digital literacy skills but does not identify the mechanism by which it was acquired in those individuals; similarly, it shows that rural older adults have interest in an educational program, but it does not identify what such a program should look like. To best address these questions, a qualitative, interview or focus group study would likely be needed, so that participants could provide full and individualized comments. Though it was not feasible to incorporate a qualitative component for this particular study due to the COVID-19 pandemic, it is undeniable that the inclusion of a qualitative component in subsequent studies will enhance the richness of the data and help contextualize the findings of the present study. Finally, there is a significant issue of the need to improve information access and technology awareness and adoption in rural communities. Recent census data has shown an exodus from rural communities across the United States (United States Census Bureau, 2021b). 92 This exodus, which is particularly driven by youth migration to larger cities, makes it even more unpalatable for political and social agencies to invest in infrastructure and education in rural communities. Rural librarians may have the most realistic path to helping abate the information and technology needs of rural older adults. How these librarians can best provide this support requires further investigation. There is a need for research that investigates what kind of specialized instruction best supports the digital literacy learning of members of the rural older adult population. Study Limitations There are several limitations of this study that deserve note. As noted in the discussion in the literature review section, there are assumptions within the methods employed in this study that should be considered as potential limitations. With the survey method, the assumption is made that the questions are sufficiently worded to be interpreted by respondents in the way intended by the researcher. The method also assumes that the questions included for the study are those that will produce meaningful data that accurately represents the population and provides a complete response to the research questions (Waugh & Subramaniam, 2018). This method generally emphasizes the perspective of a small group (sample) of individuals, which may not necessarily represent the views of the broader population. This is a weakness of most social science research in general (Choy, 2014). Additionally, there are limitations associated with cross-sectional surveys and the procurement of the sample for this study. As digital technology becomes more integrated into every facet of society, and as a new generation of adults reach older adulthood, it is likely that the adoption of digital technology and presence of digital literacy skills may increase. It would be interesting to measure data longitudinally to observe whether this phenomenon occurs—and 93 this may be possible in a future study—but this is not feasible given the time constraints of the present study. There may also be limitations in the sampling and distribution procedure, as it combines two recruitment methods: the mail survey (which used a pseudo-random selection method for recruitment—not a truly random sample) and the intermediary method (which used a convenient sampling approach). The use of intermediaries—while necessary to increase the number of respondents to the survey—introduces some possibility of response bias, as one highly-motivated intermediary might recruit a large number of participants, while a less motivated one from another county might recruit only a few, meaning that one county/group of individuals is more represented than the other. Additionally, individuals who participated at the encouragement of an intermediary may be more socially engaged in their community, since they would have to at least be socially engaged with the intermediary in order for them to share information about the survey. Responses to the questionnaire may also be skewed by social desirability bias, where they supply misleading or false answers to questions in order to respond how they believe would be most appropriate (Brenner & DeLamater, 2016). For instance, rural older adults may be hesitant to indicate that they are living in poverty, do not have a high school degree, or are less religious or politically conservative (given that these are common traits of rural populations), and so may “lie” on these questions in order to respond in a way that is more socially desirable. Similarly, respondents may attempt to provide the answers that they believe the researcher is “looking for.” While it is not possible to completely remove these biases from the study, several measures have been taken to reduce their effect in this study: the survey questions were developed from existing survey instruments that have been successfully used with the older adult population; the survey is conducted entirely anonymously; the survey is also structured so that each attribute/construct is 94 constructed based on responses to multiple questions, some of which are worded to help capture response bias if it exists (such as by flipping wording in questions, e.g., “I most like using digital technology as a source” and “I prefer to use other sources than digital technology,” where if a respondent says “yes” for the first, they should always respond “no” for the second, and vice versa). Finally, the context in which this study was conducted may limit some of the generalizability of the findings. This study was conducted during a later-stage of the COVID-19 pandemic, where many older adults have been vaccinated and are returning to a more normal way-of-life, but the threat of the pandemic is still very real. Not only does this context limit the means by which data can be collected, it could potentially influence the responses to the survey. In regard to social-relational resources, respondents may be more limited in this area during the time of the pandemic, but in the past were quite social (and may return to that more social lifestyle after the pandemic has passed). There is not much that can be done to diminish the impact on the pandemic on the findings of this study, but this, in itself, does not diminish the value of the study, as it will likely reveal evidence about rural older adult information behavior during disaster/pandemic situations that can be useful in future disaster situations. This unique context also invites the opportunity for future comparative study with the same population, which could compare behavior during and after the pandemic. Conclusion This study provided rarely studied interdisciplinary insights about older adults living in rural communities and their information behavior during the COVID-19 pandemic. A statistical path for examining these concepts was informed by Kim and Moen’s life-course model of wellbeing and Williamson’s ecological model of information behavior. The path was tested by 95 collecting data from rural older Kansans through the use of mail surveys and intermediary recruitment. The resulting 206 valid surveys were analyzed by the researcher to examine the study’s research questions. The findings of the survey support the central research hypothesis that factors of wellbeing and digital literacy are related or influence one another. Each of the three primary factors of the construct of well-being—personal resources, economic resources, and social-relational resources—was found to relate to the construct of digital literacy, itself informed by participants’ scores on an information literacy scale and trust in various information sources. Several factors, in particular, were found to correlate strongly with digital literacy skills, including age, health, sense of life control, employment and financial situation, and sociability and connectedness. Understanding the impact of factors of rural older adults may be helpful in identifying and responding to the needs of individuals most at-risk for making meaningful use of digital information. 102 Donohue, J. M., Huskamp, H. A., Wilson, I. b., & Weissman, J. (2009). Who do older adults trust to provide information about prescription drugs? American Journal of Geriatric Pharmacotherapy, 7(2), 105-116. Dou, X., Walden, J. A., Lee, S., & Lee, J. Y. (2012). Does source matter? Examining source effects in online product reviews. Computer in Human Behavior, 28(5), 1555-1563. El Skarpa, P., & Garoufallou, E. (2021). Information seeking behavior and COVID-19 pandemic: A snapshot of young, middle aged and senior individuals in Greece. International Journal of Medical Informatics, 150, article 104465. Elliot, A. J., Mooney, C. J., Douthit, K. Z., & Lynch, M. F. (2014). Predictors of older adults’ technology use and its relationship to depressive symptoms and well-being. Journals of Gerontology: Series B, 69(5), 667-677. Erdelez, S. (1999). Information encountering: It’s more than just bumping into information. Bulletin of the American Society for Information Science and Technology, 25(3), 26-29. Eriksson-Backa, K. (2008). Access to health information: Perceptions of barriers among elderly in a language minority. Information Research, 13(4), 1-12. Ferro, E., Caroleo, B., Cantamessa, M., & Leo, M. (2010). ICT diffusion in an aging society: A scenario analysis. International Conference on Electronic Government, 2010, 263-274. Gastil, J., & Marriott, R. (2019). How communicating misleading information dilutes public understanding of weight loss mechanisms. Health Communication, 34(12), 1524-1532. Gilly, M. C., & Zeithaml, V. A. (1985). The elderly consumer and adoption of technologies. Journal of Consumer Research, 12(3), 353-357. 103 Goetz, S. J., Partridge, M. D., & Stephens, H. M. (2018). The economic status of rural American in the President Trump era and beyond. Applied Economic Perspectives and Policy, 40(1), 97-118. Gonzalez, A., Ramirez, M. P., Viadel, V. (2012). Attitudes of the elderly toward information and communications technologies. Educational Gerontology, 38(9), 585-594. Goodman, R. I. (1992). The selection of communication channels by the elderly to obtain information. Educational Gerontology, 18(7), 701-714. Gordon, S. (2020). Ageism and age discrimination in the family: Applying an intergenerational critical consciousness approach. Clinical Social Work Journal, 48, 169-178. Gostin, L. O. (2018). Language, science, and politics: The politicization of public health. Journal of the American Medical Association, 319(6), 541-542. Guess, A., Nagler, J., & Tucker, J. (2019). Less than you think: Prevalence and predictors of fake news dissemination on Facebook. Science Advances, 5(1), eaau4586. https://doi.org/10.1126/sciadv.aau4586 Hales-Mabry, C. (1985). How should the information needs of the aging be met? A Delphi response. The Gerontologist, 25(2), 172-176. Hales-Mabry, C. (1993). The world of the aging: Information needs and choices. Chicago, IL: American Library Association. Hallows, K.M. (2013). Health information literacy and the elderly: Has the internet had an impact? The Serials Librarian, 65(1), 39-55. Hamm, M. E., Brown, P. J., Karp, J. F., Lenard, E., Cameron, F., …, Lenze, E. J. (2020). Experiences of American older adults with pre-existing depression during the beginnings 104 of the COVID-19 pandemic: A multicity, mixed-methods study. The American Journal of Geriatric Psychiatry, 28(9), 924-932. Hand, C., Retrum, J., Ware, G., Iwasaki, P., Moaalii, G., & Main, D. S. (2017). Understanding social isolation among urban aging adults: Informing occupation-based approaches. Occupation, Participation, and Health, 37(4), 188-198. Hargittai, E., & Dobransky, K. (2017). Old dogs, new clicks: Digital inequality in skills and uses among older adults. Canadian Journal of Communication, 42(2), 195-212. Hargittai, E., Piper, A. M., & Morris, M. R. (2018). From Internet access to Internet skills: Digital inequality among older adults. Universal Access in the Information Society, 18, 881-890. Hasan, H., & Linger, H. (2016). Enhancing the wellbeing of the elderly: Social use of digital technologies in aged care. Educational Gerontology, 42(11), 749-757. Hatcher, W. (2019). President Trump and healthcare: A content analysis of misleading statements. Journal of Public Health. https://doi.org/10.1093/pubmed/fdz176 Hawton, A., Green, C., Dickens, A. P., Richards, S. H., Taylor, R. S., Edwards, R., ... & Campbell, J. L. (2011). The impact of social isolation on the health status and healthrelated quality of life of older people. Quality of Life Research, 20(1), 57-67. Henning-Smith, C. (2020). The unique impact of COVID-19 on older adults in rural areas. Journal of Aging and Social Policy, 32(4/5), 396-402. Hill, R., Betts, L. R., & Gardner, S. E. (2015). Older adults’ experiences and perceptions of digital technology: (Dis)empowerment, wellbeing, and inclusion. Computers in Human Behavior, 48, 415-423. 105 Hodge, H., Carson, D., Carson, D., Newman, L., & Garrett, J. (2017). Using Internet technologies in rural communities to access services. Journal of Rural Studies, 54, 469478. Hollifield, C. A., & Donnermeyer, J. F. (2003). Creating demand: Influencing information technology diffusion in rural communities. Government Information Quarterly, 20(2), 135-150. Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10, 227-237. Huang, R., Kim, H., & Kim, J. (2013). Social capital in QQ China: Impacts on virtual engagement of information seeking, interaction sharing, knowledge creating, and purchasing intention. Journal of Marketing Management, 29(3/4), 292-316. Jacoby, L. L. (1999). Deceiving the elderly: Effects of accessibility bias in cued-recall performance. Cognitive Neuropsychology, 16(3-5), 417-436. Jenkins, R. (2014). Social identity. London, UK: Routledge. Jerit, J., & Barabas, J. (2006). Bankrupt rhetoric: How misleading information affects knowledge about social security. Public Opinion Quarterly, 70(3), 278-303. Johnson, K. J., & Mutchler, J. E. (2014). The emergence of a positive gerontology: From disengagement to social involvement. The Gerontologist, 54(1), 93-100. Jones-Jang, S. M., Mortensen, T., & Liu, J. (2021). Does media literacy help identification of fake news? Information literacy helps, but other literacies don’t. American Behavioral Scientist, 65(2), 371-388. https://doi.org/10.1177/0002764219869406 106 Kansas Department for Aging and Disability Services. (2020). Kansas State Plan on Aging. Retrieved from https://www.kdads.ks.gov/docs/default-source/commissions/kansas-stateplan-on-aging-fy-2018-2021.pdf?sfvrsn=e405ee_0 Kelloway, E. K. (1995). Structural equation modelling in perspective. Journal of Organizational Behavior, 16(3), 215-224. Kim, J. E., & Moen, P. (2002). Retirement transitions, gender, and psychological well-being: A life-course, ecological model. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 57(3), P212-P222. Kim, Y., Chen, H., & Zuniga, H. (2013). Stumbling upon news on the Internet: Effects of incidental news exposure and relative entertainment use on political engagement. Computers in Human Behavior, 29(6), 2607-2614. Klemmack, D. L., & Roff, L. L. (1984). Fear of personal aging and subjective well-being in later life. Journal of Gerontology, 39(6), 756-758. Komatsu, H., Yagasaki, K., Kida, H., Eguchi, Y., & Niimura, H. (2018). Preparing for a paradigm shift in aging populations: Listen to the oldest old. International Journal of Qualitative Studies on Health and Wellbeing, 13(1), e1511768. Kricorian, K., Civen, R., & Equils, O. (2021). COVID-19 vaccine hesitancy: Misinformation and perceptions of vaccine safety. Human Vaccines and Immunotherapeutics. https://doi.org/10.1080/21645515.2021.1950504 Lachman, M. E., Neupert, S. D., & Agrigoroaei, S. (2011). The relevance of control beliefs for health and aging (p. 175-190). In Handbook of the Psychology of Aging. Cambridge, MA: Academic Press. 107 Lam, K., Lu, A. D., Shi, Y., & Covinsky, K. E. (2020). Assessing telemedicine unreadiness among older adults in the United States during the COVID-19 pandemic. JAMA Internal Medicine, 180(10), 1389-1391. Lazarus, J., Ratzan, S., Palayew, A., Gostin, L., Larson, H., Rabin, K., Kimball, S., & ElMohandes, A. (2020). A global survey of potential acceptance of COVID-19 vaccine. Nature Medicine, 27, 225-228. Leibovitz, T., Shamblaw, A., Rumas, R., & Best, M. (2021). COVID-19 conspiracy beliefs: Relations with anxiety, quality of life, and schemas. Personality and Individual Differences, 175, article 110704. Levitan, L., & Wronski, J. (2014). Social context and information seeking: Examining the effects of network attitudinal composition on engagement with political information. Political Behavior, 36(4), 793-816. Lindquist, L. A., Ramirez-Zohfeld, V., Forcucci, C., Sunkara, P., & Cameron, K. A. (2018). Overcoming reluctance to accept home-based support from an older adult perspective. Journal of the American Geriatrics Society, 66(9), 1796-1799. Lockhart, R. S., & Craik, F. (1990). Levels of processing: A retrospective commentary on a framework for memory research. Canadian Journal of Psychology, 44(1), 87-112. Lund, B. D. (2021). The questionnaire method is systems research: An overview of sample sizes, response rates and statistical approaches utilized in studies. Vine Journal of Information and Knowledge Management Systems. https://doi.org/10.1108/VJIKMS-08-2020-0156 Lund, B. D. (2022). Predictors of use of digital technology for communication among older adults: Analysis of data from the health and retirement study. Working with Older People, manuscript accepted for publication. 108 Lund, B. D., & Long, B. (2021). Factors Contributing to COVID-19 Skepticism and Information Gaps Among Older Adults in the United States and Canada: An Analysis of Nationality, Gender, Education, Family, and Politics. Manuscript accepted for publication. Lund, B. D., & Ma, J. (2021). Exploring Information Seeking of Rural Older Adults During the COVID-19 Pandemic. Aslib Journal of Information Management, accepted. Lupia, A. (1994). Shortcuts versus encyclopedias: Information and voting behavior in California insurance reform elections. The American Political Science Review, 88(1), 63-76. Malbon, J. (2013). Taking fake online consumer reviews seriously. Journal of Consumer Policy, 36, 139-157. Manstead, A. S. (2018). The psychology of social class: How socioeconomic status impacts thought, feelings, and behavior. British Journal of Social Psychology, 57(2), 267-291. Marshall, C., & Rossman, G. (2006). Designing qualitative research. Thousand Oaks, CA: Sage Publications. McIntosh, W. A., Fletcher, R. D., Kubena, K. S., & Landmann, W. A. (1995). Factors associated with sources of influence/information in reducing red meat by elderly subjects. Appetite, 24(3), 219-230. McKee, S. C., Ostrander, I., & Hood, M. V. (2017). Out of step and out of touch: The matter with Kansas in the 2014 midterm election. The Forum, 15(2), 291-312. McKenzie, P. J. (2003). A model of information practices in accounts of everyday-life information seeking. Journal of Documentation, 59(1), 19-40. \ Miller, J. M. (2020). Psychological, political, and situational factors combine to boost COVID19 conspiracy theory beliefs. Canadian Journal of Political Science, 53, 327-334. Mirowsky, J. (1995). Age and the sense of control. Social Psychology Quarterly, 58(1), 31-43. 109 Moen, P., Kim, J. E., & Hofmeister, H. (2001). Couples’ work/retirement transitions, gender, and marital quality. Social Psychology Quarterly, 64, 55-71. Monteith, L. L., Holliday, R., Brown, T. L., Brenner, L. A., & Mohatt, N. V. (2020). Preventing suicide in rural communities during the COVID-19 pandemic. Journal of Rural Health. https://doi.org/10.1111/jrh.12448 Moore, R. C., & Hancock, J. T. (2020). Older adults, social technologies, and the coronavirus pandemic: Challenges, strengths, and strategies for support. Social Media and Society, 6(3). https://doi.org/10.1177//2056305120948162 Morris, A. (2007). E-literacy and the grey digital divide: A review with recommendations. Journal of Information Literacy, 1(3), 13-28. Morris, D. S., Morris, J. S., & Francia, P. L. (2020). A fake news inoculation? Fact checkers, partisan identification, and the power of misinformation. Politics, Groups, and Identities, 8(5), 986-1005. Motta, M. (2021). Republicans, not Democrats, are more likely to endorse anti-vaccine misinformation. American Politics Research. Retrieved from https://doi.org/10.1177/1532673X211022639 Murray, E., Lo, B., Pollack, L., Donelan, K., Catania, J., White, M., Zapert, K., & Turner, R. (2003). The impact of health information on the Internet on the physician-patient relationship. Archives of Internal Medicine, 163, 1727-1734. Nelson, P. (1970). Information and consumer behavior. Journal of Political Economy, 78(2), 311-329. 110 Neves, B. B., Franz, R., Judges, R., Beerman, C., & Baecker, R. (2019). Can digital technology enhance social connectedness among older adults? Journal of Applied Gerontology, 38(1), 49-72. Ng, E. (2020). Unmasking masculinity: Considering gender, science, and nation in responses to COVID-19. Feminist Studies, 46(3), 694-703. Norris-Baker, C., & Scheidt, R. J. (2005). On community as home: Places that endure in rural Kansas. In Rowles, G. D., & Chaudhury, H., Home and identity in late life: International perspectives. New York, NY: Springer. Padilla-Gongora, D., Lopez-Liria, R., Diaz-Lopez, M. P., Aguilar-Parra, J. M., Vargas-Munoz, M. E., & Rocamora-Perez, P. (2017). Habits of the elderly regarding access to new information and communication technologies. Procedia, 237, 1412-1417. Palsdottir, A. (2011). Icelanders and trust in the Internet as a source of health and life-style information. Information Research, 16(1), paper 470. Palsdottir, A. (2012a). Elderly peoples’ information behavior: Accepting support from relatives. Libri, 62(2), 135-144. Palsdottir, A. (2012b). Relatives as supporters of elderly peoples’ information behavior. Information Research, 17(4). Retrieved from http://www.informationr.net/ir/174/paper546.html#.Xbnk4ehKjcu Parente, S. L., & Prescott, E. C. (1994). Barriers to technology adoption and development. Journal of Political Economy, 102(2), 298-321. Pennycook, G., & Rand, D. G. (2019). Fighting misinformation on social media using crowdsourced judgements of news source quality. Proceedings of the National Academy of Science, 116, 2521-2526. 111 Pennycook, G., & Rand, D. G. (2020). Who falls for fake news? Journal of Personality, 88, 185200. Perrin, A., & Atske, S. (2021, April 2). 7% of American don’t use the Internet. Who are they? Pew Research Center. Retrieved from https://www.pewresearch.org/facttank/2021/04/02/7-of-americans-dont-use-the-internet-who-are-they/ Pettigrew, K. E. (2000). Lay information provision in community settings: How community health nurses disseminate human services information to the elderly. The Library Quarterly, 70(1), 47-85. Pierce, G. R., Sarason, I. G., Sarason, B. R., Solky-Butzel, J. A., & Nagle, L. C. (1997). Assessing the quality of personal relationships. Journal of Social and Personal Relationships, 14(3), 339-356. Powers, W. T. (1998). Making sense of behavior. New Canaan, CT: Benchmark Publishing. Probart, C. K., Davis, L. G., Hibbard, J. H., & Kime, R. E. (1989). Factors that influence the elderly to use traditional or nontraditional nutrition information sources. Journal of the American Dietetic Association, 89(12), 1758-1762. Quick, H. E., & Moen, P. (1998). Gender, employment, and retirement quality: A life course approach to the differential experiences of men and women. Journal of Occupational Health Psychology, 3, 44-64. Quintana, D., Cervantes, A., Saez, Y., & Isasi, P. (2018). Internet use and psychological wellbeing at advanced age: Evidence from the English Longitudinal Study of Aging. International Journal of Environmental Research on Public Health, 15(3), 480-494. 118 Williamson, K. (2005). Ecological theory of human information behavior (p. 128-132), In Fisher, K.E., Erdelez, S., & McKechnie, L.E.F., Theories of information behavior. Medford, NJ: Information Today. Williamson, K., & Asla, T. (2009). Information behavior of people in the fourth age: Implications for the conceptualization of information literacy. Library and Information Science Research, 31(2), 76-83. Williamson, K., Schauder, D., & Bow, A. (2000). Information seeking by blind and sight impaired citizens: An ecological study. Information Research, 5(4). Retrieved from http://informationr.net/ir/5-4/paper79#ref4 Wilson, P. (1977). Public knowledge, private ignorance. Westport, CT: Greenwood. Wise, K., Alhabash, S., & Park, H. (2010). Emotional responses during social information seeking on Facebook. Cyberpsychology, Behavior, and Social Networking, 13(5), 555562. Xie, B., & Jaeger, P. T. (2008). Older adults and political participation on the Internet: A crosscultural comparison of the USA and China. Journal of Cross Cultural Gerontology, 23, 1-15. Xie, B. (2020a). When going digital becomes a necessity: Ensuring older adults’ needs for information, services, and social inclusion during COVID-19. Journal of Aging and Social Policy, 32(4/5), 460-470. https://doi.org/10.1080/08959420.2020.1771237 Xie, B., He, D., Mercer, T., Wang, Y., Wu, D., Fleischmann, K. R., ... & Lee, M. K. (2020b). Global health crises are also information crises: A call to action. Journal of the Association for Information Science and Technology. https://doi.org/10.1002/asi.24357 119 Yu, T. K., Lin, M. L., & Liao, Y. K. (2017). Understanding factors influencing information communications technology adoption behavior. Computers in Human Behavior, 71, 196208. Zhao, X., Fan, J., Basnyat, I., & Hu, B. (2020). Online health information seeking using “#COVID-19 patient seeking help” on Weibo in Wuhan, China. Journal of Medical Internet Research, 22(10), article e22910. Zheng, Y., Goh, E., & Wen, J. (2020). The effect of misleading media reports about COVID-19 on Chinese tourists’ mental health. Anatolia. https://doi.org/10.1080/13032917.2020.1747208 Zhu, F., & Zhang, X. (2010). Impact on online consumer reviews on sales. Journal of Marketing, 74(2), 133-148. Zimmerman, M. S. (2021). Health information-seeking behavior in the time of COVID-19: Information horizons methodology to decipher source path during a global pandemic. Journal of Documentation. https://doi.org/10.1108/JD-01-2021-0022 120 Tables Table 1. Alignment Between Survey Questions and the Theoretical Model Concept Alignment Question Number(s) Personal Resources Age Q1 Gender Q2 Health Q19, Q20 Life Control Q8 – Q12 Economic Resources Employment Q4 Finances Q21 Device Ownership Q6, Q7 Social-Relational Resources Family/Friends Relationships Q3, Q5 Participation in Social Activities Q22 – Q29 Social Connectedness Q13 – Q18 Digital Literacy Information Literacy Scales Q38 – Q46 Trust in Information Sources Q30 – Q37 121 Table 2. Overview of Respondent Demographics Category Percentage Age 60-70 79% 71-80 17% 81+ 4% Gender Female 54% Male 46% Relationship Status Married 61% Single 24% Dating/In-Relationship 8% Widowed 8% Employment Status Full-Time 48% Not Employed 36% Part-Time 15% Looking for Employment 1% Number of Close Friends 5-9 47% 0-4 27% 10-14 14% 15+ 12% Own a Smartphone? Yes 73% No 27% Used the Internet to Seek Information? Yes 86% No 14% 122 Table 3. Response Distribution for Each Variable for Scale-Type Questions (5 is “Best”) Variable 1 2 3 4 5 Sense of Life Control 3% 17% 13% 20% 47% Social Connectedness 2% 21% 56% 21% 2% Health (Physical & Mental) 3% 23% 50% 23% 1% Finances 3% 14% 28% 29% 24% 123 Table 4. Participation in Social and Nonsocial Activities Category Never Rarely Monthly Weekly Daily Social Activities Participate in Group Activities 15% 28% 40% 10% 8% Meetup In-Person 2% 9% 53% 24% 12% Speak on Phone 0% 15% 30% 33% 22% Communicate on Social Media 14% 12% 39% 25% 9% Average for Social Activities 8% 16% 40% 23% 13% Nonsocial Activities Read Books/Magazine, Do Puzzles 5% 11% 29% 14% 41% Browse Internet 4% 5% 15% 13% 62% Average for Nonsocial Activities 5% 8% 22% 14% 51% 124 Table 5. Trust in Information Sources Category 1 2 3 4 5 Low Trust (1 + 2) High Trust (4 + 5) Difference Interpersonal Family 1% 5% 15% 52% 27% 6% 79% +73 Friends 2% 6% 28% 46% 18% 8% 64% +56 Strangers 15% 29% 37% 14% 5% 44% 19% -25 Average Interpersonal 6% 13% 27% 37% 17% 19% 54% +35 Mass Media Politicians 27% 24% 23% 10% 6% 51% 16% -35 Newspapers 12% 33% 28% 18% 9% 45% 27% -18 Social Media 22% 36% 23% 13% 5% 58% 18% -40 TV News 11% 25% 33% 21% 10% 36% 31% -5 Experts/Scientists 5% 19% 24% 33% 19% 24% 52% +29 Average Mass Media 15% 27% 26% 19% 10% 42% 29% -13 Average All Sources 12% 22% 26% 26% 12% 34% 38% +4 125 Table 6. Scores on Information Literacy Scales Score Percentage of Respondents in Each Category (Rounded Average of 5 Questions) Percentage of Respondents in Each Category Age < 70 Percentage of Respondents in Each Category Age > 70 5 0% 0% 0% 4 7% 7% 3% 3 54% 60% 41% 2 35% 31% 49% 1 4% 2% 7% 126 Table 7. Statistically Significant Findings From Correlation Analysis Variable (Higher Value) Variables Correlated With Positive/Negative (Coefficient) p-value Age (Range) Participation in Non-Social Activities Positive (.14) .04 Sense of Life Control Negative (-.14) .04 Employment Status Negative (-.41) <.01 Information Literacy Scores Negative (-.2) <.01 Trust in Interpersonal Information Sources Negative (-.24) <.01 Gender (Female) None -- -- Health (Excellent) Sense of Life Control Positive (.16) .02 Employment Status Positive (.17) .01 Finances Positive (.41) <.01 Participation in Social Activities Positive (.18) <.01 Information Literacy Scores Positive (.26) <.01 Social Connectedness Positive (.42) <.01 Trust in Interpersonal Information Sources Positive (.26) <.01 Trust in Mass Media Information Sources Positive (.21) <.01 Sense of Life Control (High) Age Negative (-.14) .04 Participation in Non-Social Activities Negative (-.42) <.01 Health Positive (.16) .02 Employment Status Positive (.19) <.01 Finances Positive (.14) .04 Family Relationships Positive (.32) <.01 Friend Relationships Positive (.2) <.01 Participation in Social Activities Positive (.18) <.01 Information Literacy Scores Positive (.2) <.01 Trust in Interpersonal Information Sources Positive (.39) <.01 Trust in Mass Media Information Sources Positive (.29) <.01 Employment Status (Employed) Age Negative (-.41) <.01 Family Relationships Negative (-.17) .01 Participation in Non-Social Activities Negative (-.37) <.01 Health Positive (.17) .01 Sense of Life Control Positive (.19) <.01 Information Literacy Scores Positive (.16) .02 127 Variable (Higher Value) Variables Correlated With Positive/Negative (Coefficient) p-value Trust in Interpersonal Information Sources Positive (.33) <.01 Finances (Able to Make Ends Meet) Health Positive (.41) <.01 Social Connectedness Positive (.38) <.01 Trust in Interpersonal Information Sources Positive (.33) <.01 Trust in Mass Media Information Sources Positive (.27) <.01 Sense of Life Control Positive (.14) .04 Information Literacy Scores Positive (.17) .01 Participation in Social Activities Positive (.22) <.01 Family Relationships (Married) Life Control Positive (.32) <.01 Employment Status Negative (-.17) .01 Participation in Non-Social Activities Negative (-.35) <.01 Social Connectedness Positive (.18) <.01 Trust in Interpersonal Information Sources Positive (.23) <.01 Friend Relationships (Many) Life Control Positive (.2) <.01 Participation in Social Activities Positive (.26) <.01 Trust in Mass Media Information Sources Positive (.14) .04 Participation in Social Activities (Frequently) Trust in Mass Media Information Sources Positive (.25) <.01 Sense of Life Control Positive (.18) <.01 Social Connectedness Positive (.36) <.01 Information Literacy Scores Positive (.22) <.01 Participation in Non-Social Activities (Frequently) Trust in Mass Media Information Sources Positive (.25) <.01 Age Positive (.14) .04 Health Negative (-.18) <.01 Employment Status Negative (-.37) <.01 Family Relationships Negative (-.35) <.01 Trust in Interpersonal Information Sources Negative (-.28) <.01 Sense of Life Control Negative (-.42) <.01 Social Connectedness Negative (-.17) .01 Information Literacy Scores Negative (-.15) .03 Social Connectedness (Very Connected) Health Positive (.42) <.01 134 Figure 3. Map of Kansas’s Congressional Districts, as of 2020 Image Credit: Institute for Policy and Social Research, University of Kansas (Copyright permission given February 9, 2021 by Xanthippe Wedel. See appendix 5.) 135 Figure 4. Map of Kansas’s 1 st with Shading Indicating the Frequency of Responses from Each County 136 Figure 5. Structural Regression Model to be Evaluated 137 Figure 6. Structural Model for this Study with Coefficients 138 Figure 7. Revised Model of Well-Being 139 Appendices Appendix 1: Introductory Letter for Survey An Invitation to Participate Emporia State University Dissertation Study Question What are the information and technology needs of older adults living in rural Kansas? Your experiences matter! Brady D. Lund, Primary Researcher Sponsored by School of Library and Information Management Emporia State University Invitation to Participate What: This survey is part of my dissertation research at Emporia State, which I am completing to earn my PhD degree. Please answer all the survey questions. It should take about 20-25 minutes to complete. All responses will be anonymous. Please return the completed survey in the provided stamped, pre-addressed envelope. 140 Who: Anyone (it need not be the addressee on the envelope) may participate in the survey who is over the age of 60 and resides in a rural community in western Kansas (west of Topeka). I am only sending out a few hundred surveys to randomly-selected households, so your responses will really have a major impact! Why: Your answers to these survey questions will help to clarify ways to better address future information and technology services to older adults who live in rural Kansas. Anonymous findings will be shared with state agencies on aging. They will also help me earn my degree! Questions: Please ask any questions you may have! Contact: Please contact me, Brady Lund, for more information by emailing me at [email protected] or you may call or send a text to 316-249-3338 for more information. I am a Ph.D. candidate at Emporia State University. Like you, I am a Kansan. I was born in Kansas and have lived in Kansas all my life. Quality of life in Kansas is very important to me and to my research supervisors. Emporia State University - Research Supervisors Mirah J. Dow, Professor Brendan Fay, Associate Professor Keith Wylie, Assistant Professor [email protected] bf[email protected] [email protected] 141 Appendix 2. List of Intermediaries for this Study Name Affiliation Contact BettyAnn Yeager Flint Hills District Extension b[email protected] Frieda Knitter River Valley District Extension [email protected] Amber Kinney River Valley District Extension [email protected] Monica Thayer Bellevue Extension mthay[email protected] Jordan Schuette Washington Extension Aging Office [email protected] Brenda Langdon Postrock Extension [email protected] Jessica Kootz Midway District Extension [email protected] Karen Shepard Twin Creeks District Extension [email protected] Kaylee Goss Walnut Creek District Extension kaylee[email protected] Jennifer LaSalle Finney County Extension jl[email protected] Melinda Daily Wallace County Extension [email protected] Friar Rich Sacred Heart Catholic Church (Colby) [email protected] Pastor Robert Alexander Trinity Lutheran Church (Colby) [email protected] Reverend Patrick Broz Colby United Methodist Church [email protected] Pastor Larry Chute North Oak Community Church (Hays) [email protected] Friar Brian Lager St. Joseph Catholic Church [email protected] Nikki Noe Dodge City Senior Center [email protected] Holy Cross Lutheran Church (Dodge City) [email protected]ail.com Adam Rankin First Missionary Church (Dodge City) [email protected] Great Bend Senior Center [email protected] Joshua Leu First Christian Church (Great Bend) [email protected] Tim Singleton First Southern Baptist Church (Great Bend) [email protected] Zach Anderson United Methodist Church (Goodland) [email protected] Dale Rosette Adventist Church (Goodland) [email protected] Ludivina Gonzalez Grant County Senior Center [email protected] Friar Juan Salas St. Anthony Catholic Church (Liberal) [email protected] First christian Church (Liberal) [email protected] 142 Pastor Brad Kefauver Crosswalk Christian Church (Minneola) [email protected] Senior Companion Program (Fort Hays State) sr[email protected] Rosie Walter Saline County Senior Center [email protected] Jami Ramsey Riley County Senior Center [email protected]m Josh Siders The Well Church (Manhattan) i[email protected] Jason Pittman First Presbyterian Church (Manhattan) [email protected] Tim Maguffee First Presbyterian Church (Junction City) [email protected] Pastor Langston Junction City Baptist Church [email protected] Golden Bell Haven (Belleville Senior Center) [email protected] Denelle Mick Cawker City Senior Center cawct[email protected] Kiowa County Senior Center [email protected] Pastor Troy First Christian Church of Logan [email protected] Cheryl Phye Pratt County Senior Center [email protected] Concordia Senior Center concordiafriendshipm[email protected] Reverend Fred Gatschet Sacred Heart Cathedral (Salina) [email protected] Craig Hauschild Trinity United Methodist Church Contact info at https://www.triumc.org/contact-us/ Melanie Hedgespeth Salina Public library [email protected] Pat Hart Norton Public Library [email protected] Sharon Springs Public Library [email protected] Lori Juhlin Dodge City Public Library l[email protected] Candis Hemel Cimarron City Library [email protected] Graham County Public Library [email protected] Atwood Public Library atwoodlib[email protected] Janice Lyhane Marysville Public Library [email protected] Steve Read McPherson Public Library [email protected] Shane Haug Frank Carlson Library (Concordia) [email protected] Debra Olds Harper Public Library [email protected] Matthew Sexton Quinter Reformed Presbyterian Church m.thomassex[email protected] Stephen Parker Christ Church Leoti [email protected] Joan Weaver Kinsley Public Library [email protected] Clay Center United Methodist Church [email protected] Audrey Flowers Meade Public Library [email protected] 143 Appendix 3: Survey Questions Instructions: Please answer each question (on both sides). When completed, please use the included envelope to mail the survey and your signed consent document back to me. Please email me at blund[email protected]a.edu or call or text me at 316-249-3338 if you have any questions. Thank you! For each of the following questions, please circle all options that apply. 1. What is your age? a. 60-70 b. 71-80 c. 81 and older 2. What is your gender? a. Male b. Female c. Other 3. Which of the following describes you? a. Married b. Widowed c. Single d. Dating/In-relationship (not married) 4. What is your current employment status? a. Employed full-time b. Employed part-time c. Looking for Employment d. Not Employed 5. How big is the circle of people you regularly interact with? a. 0-4 b. 5-9 c. 10-14 d. 15 or more 6. Which of the following devices do you own? a. Computer b. Smartphone c. Regular cell phone d. Smart technology (Alexa, Google Speaker, Ring Home Security) 7. Which of the following do you regularly use a computer or the Internet to do? a. To play games or do puzzles b. To get information (about health, news, weather, traffic) c. Watch videos, or listen to music d. Use social media (Facebook, Twitter) For each of the following questions, please select only one option that best describes you. 8. I often feel hopeless in dealing with problems in my life. a. Do Not Agree b. Agree Only a Little c. Somewhat Agree d. Agree e. Strongly Agree 9. Other people determine most of what I can and cannot do. a. Do Not Agree b. Agree Only a Little c. Somewhat Agree d. Agree e. Strongly Agree 10. What happens in my life is often beyond my control.