Participants were recruited between March and November 2023, from General and Professional Teaching Colleges (CEGEPs) and community colleges in the Montreal, Laurentian, and Bas-Saint-Laurent regions and from all universities in the province of Quebec (Canada)¹. Various methods were used to ensure that the data collection was as effective as possible: email invitations were sent to students via the institutions' mailing lists, the information was posted on the institutions' televisions and social networks, the study was promoted during lunch breaks, and so on. Social media (Facebook and Instagram) and the website of the [anonymized] laboratory were also used for recruitment. The project received ethical approval from the [anonymized university name] (2023-2256) and participants had to confirm their free and informed consent before being able to access the questionnaire.

The sample consisted of 436 young adults aged 18–30 (median age = 23.09; SD = 3.16) who were enrolled in higher education in Quebec and provided valid data. The lower age limit of 18 was set to avoid the need for parental consent and to ensure participants could provide informed consent on their own. The upper age limit of 30 was selected to reflect common definitions of young adult carers (e.g., Runacres et al., 2024) and to correspond with the widely recognized developmental stage of emerging adulthood (e.g., Arnett et al., 2014). The majority were females (72.4%), but there were also 22.8% males and 4.7% transgender, non-binary, or gender-fluid students. The proportion of young adults who reported being born in Quebec was 78.4% (n = 342), and all regions of the province were represented in the sample. Over 60% of the students (n = 281; 64.4%) still lived with at least one of their parents, and 23.2% lived with both parents. The majority of the sample reported having siblings (92.2%), and 45.7% of them were firstborns. Students enrolled in university programs were over-represented in the sample (compared to 17.3% enrolled in CEGEPs). Most participants reported being enrolled as full-time students (91.7%), a status defined in Quebec as four courses per semester at CEGEP/College or 9–12 credits per semester at university. All fields of study were represented (45.8% in humanities and social sciences—including 103 in social care or education; 46.5% in science, technology, and health; and 7.8% in arts, literature, and languages). Nearly sixty percent (60.8%) of the participants did not receive any student financial aid (AFE). Overall, 71.3% of the sample reported being employed alongside their studies (n = 311).

Data were collected through an online questionnaire on the secure Lime Survey platform. It included questions about participants' sociodemographic characteristics, the health of their relatives, the help they provided, and their educational experiences.

All statistical analyses were performed using SPSS version 29.0. Participants were divided into three subgroups according to their caregiving status: young adult carers (YAC), young adults with a sick relative but without regular caregiving responsibilities (YANC), and young adults without a sick relative (YAWSR).

Preliminary descriptive analyses (means, standard deviations, frequencies, and percentages) were conducted to characterize the sample and to examine missing data. Participants with incomplete responses on a given variable were excluded listwise from the corresponding analysis but retained for analyses in which they had valid data.

Comparisons between groups on categorical variables (e.g., gender, study interruptions, type of relative with health difficulties) were carried out using Pearson's chi-square tests. Assumptions of expected cell frequencies (>5) were checked, and effect sizes were reported using Cramer's V.

Comparisons on continuous variables (e.g., SACQ subscales of academic, social, and emotional adjustment) were examined using analyses of variance (ANOVA) and a multivariate analysis of variance (MANOVA) to account for the intercorrelations among the adjustment dimensions. When a significant main effect was observed, Tukey's HSD post-hoc tests were conducted to identify pairwise group differences while controlling for Type I error.

Assumptions of normality and homogeneity of variance were verified through graphical inspection of residuals and Levene's tests. Effect sizes were reported as η² for ANOVA/MANOVA analyses. The threshold for statistical significance was set at p < 0.05 for all tests.

Finally, given the overrepresentation of female participants in our sample compared to the general higher education population, we conducted a multivariate analysis of covariance (MANCOVA) including gender as a covariate to ensure that the observed group differences were not attributable to this imbalance.

Half of the participants (n = 227, 52.9%) reported having at least one relative whose daily life was affected by a health problem or a difficulty impacting their autonomy. Of these, 122 (i.e., 53.7% of participants who reported having a dependent relative and 27.9% of the total sample) indicated that they provided care on a regular basis or reported providing care relatively frequently and were therefore considered to be young adult carers (YAC). Young adult non-carers with a sick relative (YANC) numbered 105 (24.2%) and young adults with no sick relative (YAWSR) made up the other half of the sample (n = 209, 47.9%).

Table 1 shows the sociodemographic characteristics of the YAC, YANC, and YAWSR groups. There were no significant differences between the groups except for gender and field of study. Regarding gender, the YAC group had significantly more females than the YAWSR group χ² (N = 436, ddl = 2) = 22.047, p < 0.001, V = 0.149, and the YANC group was not significantly different from either the YAC or the YAWSR. In terms of field of study, YAC and YANC were significantly more likely than YAWSR to pursue careers in social care or education χ² (N = 436, ddl = 2) = 22.047, p < 0.001, V = 0.153.

There were significant differences between carers and non-carers among participants confronted with a loved one's difficulties. YAC were significantly more likely to live with the relative in question χ² (N = 227, ddl = 1) = 7.083, p = 0.008, V = 0.177, and their close relative was more likely to be their mother χ² (N = 227, ddl = 1) = 15.015, p < 0.001, V = 0.257 or their spouse χ² (N = 227, ddl = 1) = 14.059, p < 0.001, V = 0.249 compared to YANC. YAC were more likely than YANC to report having more than one relative with difficulties, χ² (N = 227, ddl = 1) = 13.041, p < 0.001, V = 0.240, especially both parents, χ² (N = 227, ddl = 1) = 6.050, p = 0.014, V = 0.163. No differences were observed between YAC and YANC in terms of the type of difficulties experienced by the relative or the professional help they received or did not receive.

Table 2 presents all the results related to the target variables of the study, i.e., the higher education experiences of the YAC, YANC, and YAWSR. Figures 1, 2 highlight some of the results comparing the three groups.

Regarding the “objective” indicators of school experience, the results show significant differences between the groups in terms of dropout, χ² (N = 436, ddl = 2) = 13.21, p < 0.001, V = 0.02. Orthogonal decompositions indicate that YAC were more likely (62.3%) than YAWSR (41.6%) to report interrupting or reducing their studies (Figure 1). YANC did not differ from their counterparts in the other two groups on this variable. There were no differences between the three groups on the variables “repeating” and “failing a higher education course.”

The MANOVA revealed significant differences between the groups F_{(10, 860)} = 2.406, p = 0.008, η² = 0.022 on self-report measures of academic, social, and emotional adjustment (Figure 2). This difference was found on the personal-emotional adjustment dimension, F_{(2, 433)} = 5.976, p = 0.003, η² = 0.027. On this dimension, YAC had significantly lower scores than the other two groups (−5.01, p = 0.005). The YANC scored higher than the YAC, Tukey's t = 4.11, p = 0.038, but lower than the non-carer group, Tukey's t = −3.65, p = 0.005. There were no significant differences between the three groups on the other dimensions of self-report of school adjustment (academic and social dimensions). In other words, YAC did not report lower levels of school motivation, commitment and satisfaction with academic performance or more social difficulties with their studies than their peers in the other two groups.

After controlling for gender, the differences in means between the groups persisted, according to the results of a MANCOVA. In fact, significant differences were found between groups with respect to gender, F_{(5, 406)} = 2.932, p = 0.013, η² = 0.035. Again, only the emotional adjustment variable produced significant differences, F_{(2, 410)} = 4.256, p = 0.015, η² = 0.020. Post-hoc analyses yielded the same results as above.

Consistent with international findings, YAC were found to be at a higher risk for psychological distress and emotional difficulties than their non-carer peers in higher education. This confirms the well-documented vulnerability of this population (Becker and Becker, 2008; Haugland et al., 2020). In line with the only study comparing YAC, YANC, and YAWSR (Chevrier et al., 2023a), our results show a graded pattern: YAC exhibit greater emotional vulnerability than students merely exposed to a relative's illness (YANC), who fare worse than peers without illness exposure (YAWSR). Caregiving responsibilities were also associated with more interrupted educational pathways, consistent with studies reporting delayed progression, interruptions, or constrained academic choices among student carers (Chevrier et al., 2023a,b; Day, 2021; Kettell, 2018; Haugland et al., 2020, 2022; Van der Werf et al., 2020). From a role-strain perspective (Goode, 1960), these patterns may reflect the cumulative burden of holding the resource-intensive roles of student and carer simultaneously: finite time, energy, and attention are repeatedly reallocated toward care tasks, leaving fewer reserves for studying. In addition to worrying about their loved one, juggling carer and student responsibilities can lead to frustration, guilt, fatigue and mental overload (Cass et al., 2011; Day, 2021; Kettell, 2018; Kirton et al., 2012).

Our findings reveal an apparent paradox: YAC within higher education report educational pathways marked by greater interruptions, yet they self-rate their academic commitment as comparable to that of their non-carer peers. Several mechanisms may explain this discrepancy. First, YAC may demonstrate strong resilience and bounded agency. They perceive education as a crucial avenue for developing future autonomy and identity, as well as a means of continuing to care for their loved ones (Andrewartha and Harvey, 2021; Becker and Becker, 2008; Day, 2021; Hamilton and Adamson, 2013; Haugland et al., 2022). Second, drawing on Conservation of Resources theory (Hobfoll, 1989), students likely actively reallocate resources by reducing leisure time or sleep or by relying on institutional flexibility to sustain academic engagement. While these compensatory strategies help maintain subjective involvement, they also increase the likelihood of delayed or fragmented educational trajectories. Third, self-reported measures of academic adjustment may underestimate academic difficulties because students recalibrate standards (“doing fine given the circumstances”). Indeed, studies that use “objective” measures, such as GPA (e.g., Armstrong-Carter et al., 2022) or time spent studying (e.g., Miller et al., 2024) report more academic difficulties among YAC than among their non-carer peers. However, the most significant negative outcomes (such as repeating a year or dropping out) affect only a minority of YAC enrolled in higher education, as several studies have found (Becker and Becker, 2008; Becker and Sempik, 2019; Haugland et al., 2022).

This study highlights that having a sick relative is common among students and that young carers are far from a marginal presence in postsecondary institutions. Our findings echo Stamatopoulos's (2015) description of a “hidden army of young carers in Canada.” At the same time, the proportion of young carers identified in our sample should not be considered representative of caregiving prevalence within Canadian postsecondary institutions, given the study's methodological limits. Specifically, our sample included a higher proportion of women than observed in the Quebec student population (72% vs. 55%; Linden and Jurdi-Hage, 2017) and voluntary recruitment may have attracted those more willing to disclose caregiving roles, particularly women, who are generally more likely to participate in surveys (Becker, 2022). Considering these factors, it is reasonable to assume that the prevalence of YAC in Canadian postsecondary institutions is likely to mirror recent international estimates, ranging between 12% and 18% (e.g., Chevrier et al., 2023b; Miller et al., 2024; US).

Although the recruitment design of this study does not allows for prevalence estimates, the sociodemographic characteristics of the YAC identified are similar to those reported in previous research. Consistent with prior findings, women were overrepresented among YAC (Andrewartha and Harvey, 2021; Becker and Sempik, 2019; Cass et al., 2011; Chevrier et al., 2023a; Haugland et al., 2020). A review conducted by Walker et al. (2024a) found that even in studies where men and women are equally represented within YACs, women are repeatedly overrepresented in high levels of caregiving. This result suggest both a greater involvement of women in caregiving and a greater likelihood of openly acknowledging such roles. In contrast, men may experience caregiving differently and be less inclined to disclose a relative's illness, as suggested by Chevrier et al. (2023a). Beyond gender, our data also showed that YAC were more likely to care for a mother or spouse, to have multiple relatives with health difficulties (often both parents), and to live with them. These patterns point to the influence of gender roles, physical and relational proximity, and the perceived responsibility to provide support, rather than the diagnosis itself, as key drivers of young people's caregiving roles. This interpretation is consistent with previous evidence that emotional involvement and co-residence intensify the feelings of obligation to provide care (Charles et al., 2012).

Taken together, these findings highlight caregiving as a significant yet frequently overlooked factor in students' emotional wellbeing and educational continuity. The observed pattern, comparable self-rated academic engagement alongside greater emotional strain and more fragmented trajectories, suggests that support should not be limited to cases of overt academic failure. Rather, services should treat emotional adjustment as an early indicator of risk and recognize that pauses, reduced course loads, or shifts to part-time study may be adaptive strategies rather than signs of disengagement. Furthermore, the differences seen across the three groups indicates that even students who are merely exposed to a relative's illness (YANC) also warrant light-touch preventive support, despite not meeting formal caregiving thresholds.

In practical terms, institutions can translate these principles into a few low-burden measures. First, they could embed brief, non-stigmatizing screenings for caregiving demands and recent emotional strain within routine health or advising encounters. These screenings would be paired with simple administrative indicators (e.g., credits per term, approved leaves) to detect emerging fragmentation, where consent is given. Second, create a formalized, stepped pathway coordinated by a designated contact person (e.g., a faculty YAC lead, as suggested by Kettell, 2018, or O'Keeffe, 2013), who can provide time-limited academic accommodations during acute care episodes, such as temporary deadline flexibility, access to asynchronous materials, and short-term course load adjustments, without hindering progression. Third, offer brief, transdiagnostic interventions (e.g., stress regulation, sleep and guilt management, time/energy budgeting) complemented by targeted case management to address non-medical barriers. By aligning early identification, proportionate psychological support, and narrowly scoped academic flexibility, institutions can preserve the high academic commitment that many student carers display while mitigating the emotional costs that jeopardize their academic progress.

This study provides new insights into the higher education experiences of young adult carers, but several limitations should be acknowledged. In addition to the limitations mentioned in the discussion, such as the sample not being fully representative of the Quebec student population and the use of self-reported measures only, the cross-sectional design does not allow for causal inference or examination of how caregiving responsibilities and their impact evolve over time. Longitudinal research would enable the assessment of trajectories of academic adjustment, emotional wellbeing, and persistence in higher education. This type of study is important because prior studies of postsecondary students have shown that poor mental health predicts academic difficulties, such as retakes, a decline in performance, and interruption or withdrawal (Trusty et al., 2025). This makes a “delayed” increase in school effects among YAC plausible.

This study also did not examine the variety of situations young cares may encounter. Our operationalization of caregiving was based on a threshold score of caregiving tasks, which does not fully capture the intensity, type, or duration of the support provided. Providing daily, intensive care is qualitatively different from assisting a relative occasionally, yet these variations could not be fully addressed in the present design. Additionally, this study did not allow for a systematic examination of contextual factors, such as socioeconomic status, cultural background, and the availability of institutional and community support. These factors may significantly moderate the impact of caregiving on educational outcomes. For instance, Runacres et al. (2024) demonstrate that living with a sick relative is more detrimental to the school adaptation of YAC. The possible differences in the academic experiences of YAC enrolled in distance learning vs. face-to-face courses are also worth exploring, given the controversies surrounding the benefits and challenges of distance learning for these students. Distance learning could allow YAC to more easily balance their responsibilities, but it could also deprive them of a potential personal space where they could find respite and social contact (Rawlinson, 2024). Incorporating these contextual variables in future analyses would provide a more nuanced understanding of the conditions under which YAC succeed or struggle in higher education.

Therefore, future research should build on these findings by adopting longitudinal and mixed methods designs, refining the measurement of caregiving intensity and its psychosocial correlates, and explicitly integrating contextual variables. These approaches would help identify protective factors and institutional practices that foster resilience among YAC, as well as inform policies and interventions that are tailored to their needs.

In conclusion, this study provides valuable insights into the educational experiences of YAC enrolled in higher education, while distinguishing them from YANC and YAWSR. Although YAC do not perceive their academic and social adjustment as significantly different from that of their peers, the data indicate significant difficulties in terms of disrupted academic progress and wellbeing. These findings underscore the importance of providing specific support to students who are coping with the difficulties of a loved one. This support should consider their role as carers. Such an approach will enable these students to better balance their family responsibilities and their academic goals more effectively, facilitating their transition to adulthood and the workforce.