Abstract
Background:
This article describes a protocol for a qualitative study to explore experiences and views and experiences of loneliness and social isolation in young adults, in Ireland.
Methods:
A youth Research Advisory Group will be consulted on the development of study materials. Our data collection methods will be guided by participant preferences, with potential participants choosing between virtual interviews, focus groups, and an online survey including the open-ended questions contained in the interview/focus group schedule. The data will be analyzed using framework analysis.
Discussion:
This protocol highlights methodological considerations relevant to a qualitative research study co-designed with young adults. Study findings will be relevant to mental health policy makers and other stakeholders involved in youth mental health.
Keywords
Background
Loneliness has been defined as the feeling of distress accompanying the perception that the quantity or quality of one’s social relationships is inadequate (Perlman & Peplau, 1981; Rokach, 2012). Loneliness is a fundamentally subjective experience and is distinct from social isolation in that a person can have many social ties, and still experience loneliness, or have a relatively small number of ties and only rarely experience loneliness. Although loneliness is often considered in the context of older adulthood, research indicates that rates of loneliness in young and older adulthood are comparable (Victor & Yang, 2012), and that depending on how it is measured, rates of loneliness in young adults (YA) may be as high as 71% (Qualter et al., 2015). Loneliness is an important public health issue in itself, but may also predict the development of depression (e.g., Qualter et al., 2010; van Winkel et al., 2017), as well as physical health outcomes like pain and fatigue (e.g., Jaremka et al., 2014), and mortality (O’Súilleabháin et al., 2019).
Research on loneliness across the lifespan has accelerated during the COVID19 pandemic. Several large-scale survey studies confirm that YA are particularly vulnerable to loneliness during COVID19; for example, an EU-wide survey in April 2020 reported that about 20% of adults under 35 years felt lonely all or most of the time over the 2-week period prior to interview (Eurofound, 2020). Data from the U.K. demonstrated a similar pattern; while 30% of participants across the sample (N = 5,260) felt lonely “always” or “often” in the past 7 days, the prevalence for the youngest adult group (aged 16–24) was higher at 51% (ONS, 2020). Taken together, the findings demonstrate that loneliness is a significant public mental health issue during the young adulthood life stage.
Young adulthood is difficult to define and varies to some degree across cultures. Until the 1960s, the transition to adulthood for most social class groups in industrialized western countries was relatively well defined by events like marriage and parenthood (Eccles et al., 2003). However, with greater heterogeneity in the activities of this life stage, we no longer have a “small, easily understood, set of patterns for the transition to adulthood” (Eccles et al., 2003, p. 384). This has lead researchers to describe emerging adulthood as a distinct life stage spanning from approximately age 18–25 (Arnett, 2000), involving demographic, social, and psychological changes, including identity exploration, shifts in the relative importance of relationships of one’s family and peers, and important decisions about education and occupation. Thus, emerging adulthood can be viewed as a time of significant change or potential for change in both structural and functional dimensions of a person’s wider social networks and close relationships, besides other changes in demographic, social, cognitive, and biological factors that might influence loneliness. However, relatively little is known about the experience of loneliness during this life stage.
The existing qualitative literature has focused on the experience of loneliness for YA at university (Vasileiou et al., 2019), or in depression (Achterbergh et al., 2020). The findings suggest that there are several reciprocal pathways between loneliness and depression in young adulthood, including social withdrawal due to poor mental health, and due to reluctance to disclose depression (Achterbergh et al., 2020), and that YA experiencing loneliness after moving for university use a wide range of coping strategies including support-seeking (Vasileiou et al., 2019). Besides this, while YA are included in studies of loneliness in particular groups like those with rheumatoid arthritis (Bay et al., 2020), little qualitative research has directly explored loneliness during emerging adulthood.
Theoretical Positioning
This study draws on macro-theories of development that describe “emerging” adulthood as a life stage that is theoretically and empirically distinct from adolescence and fully-fledged adulthood (Arnett, 2000). This study is also informed by several theories of loneliness that view loneliness as a personal characteristic (e.g., the cognitive discrepancy model; Perlman & Peplau, 1982), as an interaction between the individual and the environment (e.g., the social needs approach; Weiss, 1973), or as a product of the wider society (e.g., the idea of “liquid modernity”; Bauman, 2000). Theoretical approaches to loneliness are not mutually exclusive and loneliness is likely to be influenced by multiple levels of variables; therefore, rather than ground this research in one theoretical perspective, items tapping into several theories are included in our data collection materials (see Supplemental Files 1 [interview schedule], 2 [focus group schedule], and 3 [demographic items for all participants, and survey schedule]).
Importantly, our aim is to understand loneliness from the perspective of young adults themselves; thus, while it would be wasteful not to use existing theories of loneliness to inform our schedule, we aim to prioritize young adults’ descriptions and narration on this topic, regardless of whether their perspectives align with a particular theory, or not.
The Present Study
In summary, given relatively high prevalence of loneliness in emerging adulthood, it is important to understand the nature of the experience of loneliness and young adults’ views on factors that precipitate or maintain loneliness. This article describes the rationale and methodology for a qualitative study to explore loneliness and social isolation in young adulthood. This qualitative study invites YA to share their views and experiences of loneliness to (1) better understand the experience of loneliness from the perspective of young adults and (2) identify what young adults believe precipitates and maintains feelings of loneliness. Unlike previous studies, this study will provide an in-depth understanding of loneliness in particular, rather than loneliness as an adjunct to the related experience of depression, or in the context of a particular life transition. The study materials are designed in consultation with a YA Research Advisory Group recruited via Ireland’s youth information website SpunOut.ie, and multiple forms of data collection will be offered to maximize opportunities for YA to participate. Findings will provide valuable insights into the experience of loneliness from the perspective of YA, and a basis for quantitative research investigating risk factors for loneliness in this group.
Explanation and Justification of Method
Design: Semi-Structured Interviews, Focus Groups, and Survey
Participants will be recruited via advertisements placed on social media, on the SpunOut.ie website, via email to local universities, secondary schools, and via community organizations representing the participant groups of interest. Participation will be incentivized via optional entry into a prize draw for all participants, with an additional one-4-all voucher for all interview and focus group participants. Participants can choose to participate in a semi-structured, virtual interview or focus group (via Zoom or MS Teams), or to complete an online survey (via Qualtrics). Our preferred data collection method is semi-structured virtual interviews. The rationale for also offering small focus groups (to include approximately three to five participants) is based on preliminary guidance from SpunOut.ie, indicating that some YA subgroups prefer to discuss ideas in group settings. Although interviews allow us to gain detailed insights into individuals’ views and experiences, facilitating focus groups may allow us to include subgroups that may be reluctant to participate in interviews. Finally, YA who indicate their willingness to participate, but who prefer not to participate in either an interview or focus group, will be invited to complete an online survey containing core items relating to loneliness as well as a random subset of the other topic areas included in the interview schedule. The rationale for offering this third method of data collection is that (a) this further increases the likelihood that individuals from under-represented groups can be included in the study and (b) given the COVID pandemic, some YA may not be able to privately participate in a virtual interview or focus group.
The interview and focus group schedules (see Supplementary Files 1 and 2) will contain a set of open-ended questions based on the preliminary findings of an on-going scoping review of the literature (see Kirwan et al., 2021), and items suggested by our Research Advisory Group. The interview schedule may be modified to suit participant subgroups (e.g., different examples of life transitions may be used for different groups). The purpose of the schedule will be to ensure all questions are covered; however, open narration on participants’ perspectives and experiences will be encouraged. Interviews and focus group discussions will be audio-recorded and transcribed verbatim. Written informed consent will be obtained from study participants prior to participation.
Sampling/Recruitment
Our aim is to understand the experience of loneliness from the perspective of young adults, using an interview schedule co-designed with young adults (the Research Advisory Group). SpunOut.ie is a youth information website and the Research Advisory Group is drawn from members of SpunOut.ie’s Youth Action Panels. The panels represent YA from across Ireland and include an LGBTQI panel, a BAME (Black Asian Minority Ethnic) panel, a Traveler panel (members of the Irish Traveler community, an indigenous ethic minority group who traditionally lived nomadically), and plans to develop a panel of YA with experience of direct provision (Ireland’s current reception system for asylum seekers), are underway. Research Advisory Group members were drawn from a group who responded to a notice placed on the panels’ internal communications system.
We aim to collect data from 30 YA who represent the diversity of the SpunOut.ie panels. Therefore, we aim to collect data with YA from the LGBTQI community (minimum target n = 5), people with disabilities (minimum target n = 5), and minoritized groups including the Traveler community (minimum target n = 5), as well as from the wider community (n = 15) and to include a gender and urban/rural mix in the sample. If our efforts to recruit minoritized groups are successful, we will increase our target sample size and endeavor to include a Traveler-specific subgroup (minimum target n = 5) in our research.
Importantly, the perspectives of the Research Advisory Group and wider Youth Action Panels will determine whether we collect data with these subgroups. Members of the Research Advisory Group are viewed as representatives of their respective communities. If these representatives believe it would not be beneficial for their communities to participate in the study, we will not purposively recruit from those specific communities. To be clear, we are not anticipating that the stakeholders will view the study as problematic. It is more likely that our demographic measures, interview schedule and mode of data collection will be modified to account for different community preferences. However, if there is any sense that studying loneliness in these particular communities reinforces stereotypes or meta-stereotypes relating to a particular group, we will not pursue recruitment from these particular communities.
Data Handling/Analysis
The audio recordings of the data will be transcribed verbatim; when these are verified by the research team the audio recordings will be destroyed. Data will be analyzed using NVIVO. Following study completion, the anonymized transcripts will be deposited in the Irish Qualitative Data Archive (IQDA; subject to participant consent). The protocol has also been pre-registered on osf.io (Creaven et al., 2021).
Deciding on an Analytic Approach
Framework analysis (Ritchie & Spencer, 1994) will be used to analyze the data. This decision has been guided by our primary research aim of understanding views and experiences of loneliness from the perspective of young adults (i.e., without being constrained by a single particular theoretical account of loneliness). This research aim aligns with an analytic approach with an explicit focus on experience, as well as the context in which the experience occurs.
Before deciding on framework analysis, a number of possible approaches were considered. One such approach is interpretive phenomenological analysis (IPA). This approach aims to understand people’s experience of a particular phenomenon and how they make sense of the experience (Smith & Shinebourne, 2012). Its unit of analysis is the group of individuals who have similar experiences of a phenomenon, and a focus on each individual’s unique experience is a core feature of this approach. Because we are interested in the experience of particular subsets of young adults that have been under-represented in previous research and that may not share similar experiences, this approach is not optimal for our study.
Thematic analysis (TA; Braun & Clarke, 2020) was also considered. Thematic analysis refers to a cluster of approaches that may have different underlying philosophies but share a focus on capturing patterns in data. Framework analysis was considered preferable to TA, because the former emphasizes how both a priori concerns and data driven themes should guide the development of the analytic framework. This was something that aligned with our goal, in that we wanted to draw on the extensive theoretical accounts of loneliness and of young adulthood for the development of our interview schedule, while prioritizing the actual views and experiences of our participants, whether these related to a priori concerns, or not.
Importantly, framework analysis is not bound by a particular epistemological position (Ritchie & Spencer, 1994). The defining feature of framework analysis is the matrix output: rows (cases), columns (codes) and “cells” of summarized data, providing a structure into which we can systematically reduce the data, in order to analyze it by case (i.e., individual interviewee) and by code (Gale et al., 2013). The determinants of loneliness are multifactorial and individual experiences of loneliness may have different perceived causes and consequences. Framework analysis can allow us to explore similarities across participants in our sample, while also understanding how different categories of themes relate to individual participant experiences.
Parkinson et al. (2016) cite examples of each of the four types of research question Ritchie and Spencer (1994) outline as questions that can be addressed by FA. These are contextual (identifying the form and nature of what exists (e.g., exploring the experience of pregnant women suffering from psychological distress; Furber et al., 2009), diagnostic (examining the reasons for, or causes of, what exists (e.g., exploring the barriers to seeking help in young women with eating disorders; Evans et al., 2011), evaluative: appraising the effectiveness of what exists (e.g., evaluating the helpful factors of group interventions for anxiety and depression, and what improvements can be made; Newbold et al., 2013), and strategic: identifying new theories, policies, plans or actions (e.g., identifying the requirements for implementing pediatric care closer to home; Heath et al., 2012). Our goals are contextual (exploring the experience of loneliness in young adulthood) and diagnostic (examining the perceived reasons for, or causes, of loneliness), indicating that FA may be a useful technique for our primary research questions. However, depending on the mix of interview, focus group, and questionnaire data we succeed in collecting, this approach might be applied only to the “thick” interview and focus group data, with content or thematic analysis applied to the questionnaire data. We also anticipate that other analytic approaches may be used for secondary analyses of the data, in future.
Framework Analysis Process
Framework analysis involves five stages: familiarization, identifying a framework, indexing, charting, and mapping and interpretation (Ritchie & Spencer, 1994). Familiarization involves “getting to know” the data and is a characteristic of almost all qualitative approaches. For FA, it is not necessary to review all material at this stage. For ELSIE, the interviews will be conducted by a small team of research assistants, and the analysis team (which will overlap at least in part with the interview team), will need to develop familiarity with the data. For this process, we plan to (1) listen to and read transcripts representing a diversity of views from participants (e.g., including a mix of genders), and (2) discuss our sense of the data with the interview team.
The second stage, identifying a framework, involves managing and organizing the data. The qualitative data analysis software package NVivo will be used to facilitate data management (Bazeley & Jackson, 2015). Parkinson et al. (2016) advise that the distinction between this stage and the final “mapping and interpretation” stage requires emphasis, as thinking about the data in an interpretative way to begin with, obstructed the development of their framework. Based on their guidance, we will (1) use a priori concerns from our interview schedule to guide category development, and (2) test these categories on a proportion of the data set with the aim of revising them, in light of data-driven issues. Based on prior research, we will also include categories for “interaction between interviewee and interviewer,” and “other”; if it transpires that we have a range of interesting data categorized as “other,” we may revise our framework.
The third stage, indexing, involves organizing the transcripts into the framework categories by systematically applying the framework to each interview transcript. At minimum two members of the team (EK and AMC) will take responsibility for indexing and charting the data. Each will work through the transcript text, highlighting a chunk of the text and deciding which category (or categories) from the framework to assign the text to. During the indexing stage, we may encounter chunks of data that do not fit neatly into one category. However, as indexing is intended to make a dataset more manageable, we will discuss whether there are instances when a particular chunk of text can be coded into multiple categories.
The fourth stage, charting, involves summarizing the indexed data for each category and organizing the summaries in chart form. This structure allows the summaries to be read for participant-level analyses, or for the analysis of a specific theme or category. To chart an interview, we will work through each framework category, summarizing all data that have been indexed to that category, providing a summary for each category, for each participant. We aim to keep the summaries concise to make the dataset manageable.
The fifth and final stage, mapping and interpretation, involves moving beyond managing the data and focuses on making sense of the data as a whole. This stage involves identifying patterns in the data and making sense of it in light of our research questions. For our study, EK and AMC will independently review the data, discuss our impressions, and together develop a set of themes representing the data. EK and AMC will then present these preliminary themes to the other members of the research team, to question, comment, and further develop.
A strength of the FA approach is the emphasis on data management. This is appealing for our project as it aligns with our wider focus on data management in terms of making anonymized transcripts available for future use via IQDA. Further, NVivo is compatible with FA and has previously proven to be an effective tool to aid data management in the development of an analytic framework (see Bonello & Meehan, 2019; Parkinson et al., 2016). The emphasis on data management in FA is also relevant in the context of a sample with characteristics that are not shared by the research team. For example, none of the research team is a member of a minoritized group. We hope that this relatively systematic approach to data charting will aid us in capturing the perspectives of these participants.
Ethics
Key ethical concerns are (1) informed consent (2) anonymity and confidentiality (3) risk of distress.
The information sheet and consent forms will be reviewed by the Research Advisory Group to ensure these are accessible and facilitate informed consent. Written informed consent will be obtained for interviews and focus groups; participants for the online survey will indicate their consent on-screen before the survey is accessible to them. As part of the consent process, participants can allow anonymized transcripts to be archived, via IQDA, for further research in accordance with FAIR data principles. Direct identifiers will be removed; indirect identifiers will also be removed, if necessary. Only anonymized excerpts will be used for dissemination. If the data relating to a particular category of participant reveals information about that wider group that is considered sensitive, transcripts may be abbreviated, and this will be noted in the meta-data.
Loneliness may be considered a sensitive topic. It is possible that participants may experience some degree of distress describing personal experiences of loneliness, despite being willing interviewees. This is especially important for young adults who have experienced or are currently experiencing mental health issues. To minimize risk of distress, participants can choose to participate in an interview, focus group, or anonymously via the online survey. Participants need only respond to the questions they would like to respond to and will have the option to take a break or end the interview at any stage. As representatives of their young adult communities, participants can share their views on loneliness without disclosing details of personal experiences. If participants do exhibit distress, the interviewer will conclude the formal interview, thank the interviewee for their time, and move to the debriefing information. Further, the debriefing sheet will contain contact details for relevant support services.
In research, there is typically a power imbalance between the researcher and the participant (“the researched”), such that the researcher directs the process while the participant responds (Reid et al., 2018). We aim to minimize the power differential in three ways. First, we have adopted a participatory approach in which young adults will be involved in the refinement of the interview schedule, decisions about their communities’ participation, and decisions about dissemination. Second, our data collection methods will be guided by participant preferences, with potential participants choosing between virtual interviews, focus groups, and an online survey including the open-ended questions contained in the interview/focus group schedule. Finally, the interviews will be conducted by research assistants who are close in age to our target group, which we hope will enable empathetic interactions.
Dissemination
To maximize the value of YA participation in the study, we will ensure that the transcripts of their interviews are anonymized and archived for access by others, in the future. The findings from this study will be disseminated to academic audiences via journal articles and conference presentations. The findings will be disseminated to lay and YA audiences via: (1) lay summary of and infographic of findings for dissemination via SpunOut.ie (2) video animation of key study findings and (3) SpunOut.ie blogposts describing the partnership and study findings. These dissemination plans may be amended based on feedback from our YA Research Advisory Group.
Rigor
Several strategies will be adopted to ensure qualitative rigor. First, researcher reflexivity will be considered throughout the research process, by writing a reflective piece on our (the researchers) positionality at the outset, and by revisiting this at each subsequent stage (data collection, analysis, and generation of themes). Second, inter-coder reliability will be assessed in NVivo (O’Connor & Joffe, 2020). The use of NVivo also facilitates transparency during analysis as summaries or interpretations can easily be linked back to the raw data. Third, data from different sources will be contrasted to determine if different findings are observed, particularly in terms of the contrast between the questionnaire and other modes. Finally, “thick description” (Pontoretto, 2006) will be included in the final report, with examples of raw data (i.e., direct quotes from study participants) alongside important context relevant to the data (e.g., information about the participant).
Discussion
Our knowledge of the experience of loneliness in young adulthood is currently insufficient. Using a qualitative design, this study aims to provide an in-depth account of YA experiences and views of loneliness, to inform subsequent research in this area. The findings will be disseminated to both lay and academic audiences and the data will be archived for further analysis.
There are a number of research limitations. First, we aim to recruit a more diverse sample than has been included in previous studies, meaning that the extent to which interviewers and interviewees are perceived to have shared characteristics is likely to vary across participants, and may have a bearing on how participants’ experiences are discussed. Second, remote interviews and focus groups mean that we may not be able to achieve the same level of depth as might be possible with face to face research. However, remote methods may also facilitate participation from a wider range of participants and YA tend to be relatively familiar with virtual environments. Further, remote research may also allow us to draw new methodological insights.
Supplemental Material
Supplemental Material, sj-pdf-1-ijq-10.1177_16094069211028682 - Protocol for a Qualitative Study: Exploring Loneliness and Social Isolation in Emerging Adulthood (ELSIE)
Supplemental Material, sj-pdf-1-ijq-10.1177_16094069211028682 for Protocol for a Qualitative Study: Exploring Loneliness and Social Isolation in Emerging Adulthood (ELSIE) by Ann-Marie Creaven, Emma Kirwan, Annette Burns and Páraic S. O’Súilleabháin in International Journal of Qualitative Methods
Supplemental Material
Supplemental Material, sj-pdf-2-ijq-10.1177_16094069211028682 - Protocol for a Qualitative Study: Exploring Loneliness and Social Isolation in Emerging Adulthood (ELSIE)
Supplemental Material, sj-pdf-2-ijq-10.1177_16094069211028682 for Protocol for a Qualitative Study: Exploring Loneliness and Social Isolation in Emerging Adulthood (ELSIE) by Ann-Marie Creaven, Emma Kirwan, Annette Burns and Páraic S. O’Súilleabháin in International Journal of Qualitative Methods
Supplemental Material
Supplemental Material, sj-pdf-3-ijq-10.1177_16094069211028682 - Protocol for a Qualitative Study: Exploring Loneliness and Social Isolation in Emerging Adulthood (ELSIE)
Supplemental Material, sj-pdf-3-ijq-10.1177_16094069211028682 for Protocol for a Qualitative Study: Exploring Loneliness and Social Isolation in Emerging Adulthood (ELSIE) by Ann-Marie Creaven, Emma Kirwan, Annette Burns and Páraic S. O’Súilleabháin in International Journal of Qualitative Methods
Footnotes
Acknowledgments
We are grateful to Ian Power (CEO) and Sinead Beirne (Outreach & Training Officer) of SpunOut.ie for assistance recruiting our Research Advisory Group, and to the Research Advisory Group for their contribution to the study. We also acknowledge guidance from Dr. Daragh Bradshaw in relation to qualitative analysis.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project is funded by an Irish Research Council New Foundations Award to the first author (NF/2020/21017360).
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
