What does it mean to be alone? An analysis of interpretations of solitude among adolescents and adults

Introduction

Spending time alone becomes more prevalent and enjoyable during adolescence and into adulthood (Borg & Willoughby, 2021; Corsano et al., 2006; Larson, 1990). Interpretations of what it means to be alone may vary across developmental periods. For example, adolescence and early adulthood are times when forming meaningful peer relationships are important (Bowker et al., 2021; Schnittker, 2007), and thus being alone may be conflated with feeling alone if existing relationships are inadequate. Further, experiences with being alone are complex and can be experienced in different ways (e.g., being alone may be a positive experience but may become negative if it is done too often; Coplan et al., 2021). Research has investigated whether children can distinguish among concepts related to being alone (i.e., loneliness, solitude, aloneness; Galanaki, 2004), and has qualitatively assessed participants’ attitudes toward feeling alone (e.g., Hemberg et al., 2021; see Achterbergh et al., 2020 and Mansfield et al., 2021 for reviews); however, studies have not specifically asked participants to report what being alone means to them. The goal of the current study was to use qualitative data to explore three research questions: (1) How do adolescents and adults interpret being alone? (2) Do these interpretations of being alone differ between the samples? (3) Do these interpretations differ across age and sex within each sample?

Method

Participants

The adolescent sample comprised of 645 participants (M_age = 14.55, SD = 1.74, range = 11 to 19 years, 50.2% female), and the adult sample comprised of 444 participants (M_age = 36.90 years, SD = 13.84, range = 18–83 years, 69% female, 0.5% other). Both samples completed separate online surveys. See Appendix for socio-demographic information for both samples. Both studies were approved by the University Research Ethics Board. Missing data for age and sex in the adult sample (n = 6, 1.4%) were imputed using the expectation–maximization (EM) algorithm with all study variables included in the analyses to avoid the biased parameter estimates that can occur with pairwise or listwise deletion (Schafer & Graham, 2002). There were no missing data for age and sex in the adolescent sample.

Measures

Age and sex were reported in both samples. Participants in both samples were asked to provide an open-ended response to the following prompt: “Please provide a description of what being alone means to you. Your description can be as short or as long as you would like”. Responses were coded using an inductive approach to thematic analysis. The process of thematic analysis was conducted as outlined by Braun and Clarke (2006).

Data Analysis

Ambiguous responses were removed (n_adolescent = 20, n_adult = 3). Inter-rater reliability for 30% of responses was good for both the adolescent and adult samples (weighted Cohen’s kappa = 0.917 and 0.868, respectively). Discrepancies were discussed between researchers until agreement was reached. The remaining responses were coded by one researcher (the first author) based on an agreed upon dichotomous nominal scale for each theme (0 = theme not referenced in response, 1 = theme referenced in response). Responses were coded under multiple themes if applicable.

Results

Across both samples, four themes emerged. See Table 1 for descriptions of each theme, and examples and prevalence of responses in both samples. Crosstab analyses were conducted to assess differences in prevalence of each theme across samples. Binary logistic regressions were included to assess whether age and sex were associated with each theme within each sample. A Bonferroni correction was applied when assessing the significance of each model (p = 0.013).

Table 1 Examples of coded responses from adolescent and adult samples

The chi-squares for the loneliness and negative themes were significant (χ ² (1, N = 1066) = 8.63, p = 0.003 and χ ² (1, N = 1066) = 8.51, p = 0.004, respectively). Adolescents described being alone as loneliness less often than expected compared to adults who described being alone as loneliness more often than expected. Adolescents also described being alone as negative more often than expected compared to adults who described being alone as negative less often than expected. The chi-squares for physical separation and positive themes were non-significant (ps > 0.013).

The logistic regression models for loneliness were significant in both samples (ps < 0.013). Being female (OR = 1.49, p = 0.015) and higher age (OR = 1.10, p = 0.049) in the adolescent sample, and lower age (OR = 0.971, p < 0.001) in the adult sample were associated with describing being alone as loneliness. The regression model for physical separation was significant in the adolescent sample only (p < 0.013), such that being male (OR = 0.669, p = 0.018) and lower age (OR = 0.883, p = 0.012) were associated with describing being alone as physical separation. See Table 2 for parameter estimates. The regression models for positive experience and negative experience were non-significant for both samples (ps > 0.013).

Table 2 Parameter estimates for significant logistic regressions in the adolescent and adult samples

Discussion

The current study explored interpretations of being alone among adolescents and adults, whether these interpretations differed between these samples, and whether interpretations differed within each sample in relation to age and sex. Four themes emerged, indicating that there are distinct interpretations of being alone.

Age differences that emerged in interpreting being alone as loneliness appear to support past research that loneliness peaks in late adolescence (and often early adulthood), likely because there is considerable emphasis placed on social relationships (Child & Lawton, 2019). Thus, interpreting being alone as feeling alone may be more prominent during this time. Adolescents desribed being alone as negative more often than expected compared to adults, however, this may be due to adolescents reporting that being alone is “boring”. Sex differences support past research that female adolescents may be more susceptible to feeling lonely (Martel, 2013) and thus interpret being alone this way, compared to males who may interpret being alone more objectively (i.e., as physical separation).

The current study empirically explored the ways in which adolescents and adults interpret being alone using qualitative data and a large sample size. The current study used two different samples, which limited analyses to concurrent associations. Data collection took place during the coronavirus-19 pandemic, which may have affected how participants interpreted being alone (though participants made very few references to the pandemic in their responses). Nonetheless, findings suggest that interpretations of being alone are complex and highly variable among adolescents and adults and should be considered in order to better understand the implications of being alone.

Appendix

Socio-Demographic and Survey Characteristics

Adolescent Sample

The participants in the adolescent sample were students involved in a larger longitudinal study of health-risk behaviours among elementary and high school students. Given that the question about what it means to be alone was only included in the fifth wave of the larger longitudinal study, only that wave of the data is analyzed here. Participants were enrolled in several schools in Ontario, Canada, and were surveyed annually, beginning in 2017. Parent reports indicated that 69.3% of adolescents were White, 18% were Black, 1.6% were Asian, 3.9% were Hispanic, 0.3% were Indigenous, and 6.7% were Mixed (0.3% preferred not to answer). Parental education level was used as a proxy for socioeconomic status, and the prevalence was 7.8% for a professional or advanced degree, 25% for an undergraduate degree, 58.3% for an associate degree or diploma, 7.1% for some college, university, or technical diploma, and 1.9% for a high school diploma.

Students in the adolescent sample earlier were invited to participate in the study during visits to their school. Due to restrictions related to the coronavirus-19 pandemic, the wave of the survey used in the current study was administered online through Qualtrics from March–June, 2021. The entire survey took approximately one hour to complete. Students received $30 as compensation for their completion of the survey. Parents provided informed consent and participants provided informed assent before completing the survey. The survey was developed by the second author. This study was approved by the University Research Ethics Board.

Adult Sample

The participants in the adult sample were adults from the United Kingdom who were recruited to participate in an online survey through Prolific, an online data collection platform. The majority of participants reported their nationality as European (89.6%). Others reported their nationality as Asian (4.3%), African (2.0%), South American (0.7%), Australian (0.5%), North American (0.2%), and 1.6% preferred not to answer. Participants reported their highest level of education completed as a proxy for socioeconomic status, and the prevalence was 52% completed some college, university, or apprenticeship, 3% completed a college/apprenticeship or technical diploma, 25% finished high school, and less than 1% did not finish high school.

Participants in the adult sample volunteered to complete the online survey through their personal Prolific accounts. Participants indicated informed consent online before beginning the survey. The entire survey took approximately 20–30 minutes to complete. Data was collected in April, 2021. All participants who provided the correct completion code were compensated with €2.50. The survey was developed by the first and second authors. This study was approved by the University Research Ethics Board.