Canada has a rapidly growing refugee population, yet, there are limited research studies on the physical health of working-age refugees in comparison to the health of immigrants and Canadian-born individuals. Investigating social capital and acculturation measures may provide important insights into the factors associated with good self-reported health and this may help to inform health promotion strategies for refugees in Canada. A secondary analysis was conducted on data collected from the Canadian General Social Survey 27 (GSS-27) comparing a sample of refugees (n = 753), immigrants (n = 5,063), and Canadian-born (n = 11,266) respondents between the ages of 15 and 64. Both bivariate and logistic regression analyses were conducted. Self-reported physical health, dichotomized into poor versus good, was the outcome of interest. The self-reported physical health status of refugees, immigrants, and Canadian-born respondents was comparable. Visible minority status was not significantly associated with self-reported health status. Among refugees, the likelihood of reporting good health was associated with being a woman, being married/common-law, being involved in a social group/organization, and having more than half of one’s friends who spoke a different mother tongue than the respondent. Refugees, however, were less likely to have a confidant and be involved in social groups/organizations as compared to immigrants or those born in Canada. The odds of reporting good health were significantly lower among those who had experienced discrimination within the last five years. Social capital and acculturation may be protective of the self-reported health of refugees in Canada. Initiatives to support refugees’ social connections are therefore warranted.

1. Self-Reported Health of Working-Age Refugees, Immigrants, and the Canadian-Born

Refugees are individuals who are forcibly displaced from their home country because of war, violence, and fear of harm or persecution due to their group membership (e.g., race, politics, religion, and nationality) [1]. The number of refugees in Canada has been increasing in recent years, with over 31,000 refugees [2] arriving in 2019. Refugees tend to be younger than the Canadian-born population with 57% of refugees being aged 25–54, compared to 38% of those born in Canada who are in this age group [3]. The median age of refugees at their time of arrival is 23 years old. However, for some refugee groups, the median age at arrival is younger,such as Syrian refugees, who are, on average, 18 years old when they arrived in Canada [4].

There is a robust literature, however indicating that refugees have worst mental health than either immigrants or the Canadian-born [5]. Despite the growing number of refugees in Canada, there are few studies focusing on the social determinants that may affect their physical health outcomes, with the vast majority of research conducted with refugees in Canada focusing on mental health rather than other health outcomes [6]. Some studies in the US indicate that refugees are vulnerable to chronic physical health conditions. For example, research among Cambodian refugees (mean age = 59) suggests an increased risk for heart disease, which can be associated with higher levels of hypertension and cholesterol compared to the general US population [7]. Furthermore, a study of Iraqi refugees in the US (89% of the sample aged 18–59) reported that their most frequent health problems were hypertension, high cholesterol, diabetes, and obesity, with 60% of the sample having at least one of these conditions and 37% reporting two or more of these health issues [8], and, in a Canadian study, refugees (aged 15–39) had a slightly higher prevalence of hypertension compared to the Canadian average (30% vs. 23%) [9]. Refugees in the US also have a higher prevalence of stroke, arthritis, and disabling chronic pain when compared to immigrants [10]. On the other hand, a systematic review of Canadian researchers reported that refugees had better health compared to individuals born in Canada with respect to cancer and mortality from chronic diseases (e.g., heart disease, diabetes, and respiratory conditions) [11].

More than one in five Canadians is an immigrant [12]. In Canada, a number of studies have indicated that there is a “healthy immigrant effect,” with better health, on average, among recent immigrants compared to those who are Canadian-born [11, 13, 14]. However, immigrants’ health advantage diminishes with increasing residency in Canada [15, 16]. One explanation for the initial health advantage among immigrants may be due to the mandatory Immigrant Medical Exam (IME) that all newcomers must pass prior to arrival in Canada. The IME assesses immigrants for (in)admissibility based on medical conditions that may be burdensome on Canada’s health care system. In contrast, refugees who are deemed to have complex health needs are exempt from medical inadmissibility due to their refugee status [17].

Population-based research is needed to investigate whether there are health disparities between refugees, other immigrants, and those who are Canadian-born, and if so, what are some potential reasons for these differences. Social determinants of health including socioeconomic status, employment, and education may contribute to health inequities experienced by refugees. A recent report from UNHCR summarized the socioeconomic and employment trends of refugees following their arrival in Canada [3]. In terms of income, it was reported that refugees earn 50% less than those who are Canadian-born during their first year of arrival, but that at five years postsettlement, refugees have comparable salaries to their Canadian peers of $40,000–$79,999 per year (23% vs. 27%, respectively) [3]. The UNHCR report found that unemployment rates of working-age refugees were comparable to that of the Canadian-born population (9% vs. 6%). The report also concluded that many refugees were in white-collar jobs with 51% of working refugees employed in jobs that require professional training, and post-seoncdary qualifications[3]. However, racialized groups may experience unique barriers to employment, such as workplace or job selection discrimination, language barriers, and foreign training that is not recognized by Canadian employers [18]. For example, in one Toronto study with racialized refugee and immigrant women, respondents shared experiences of low-paying, insecure, and underskilled forms of employment which negatively impacted their health due to lack of access to health benefits (e.g., medication and dental coverage), long working hours, poor or unsafe working conditions, and few sick days [18].

The health of refugees in Canada may also be influenced by other postmigration factors. Some research studies have suggested that the experience of discrimination may contribute to poorer health among refugees. An Australian study with African, Middle Eastern, and Southeast Asian refugees reported that among those who experienced discrimination since their arrival in the host country, 90% stated that discrimination had a negative impact on their health [19]. One Canadian study found that refugees from Africa described experiencing discrimination on the basis of race within employment and daycare settings [20]. Although there are fewCanadian research studies on the impacts of discrimination on the physical health outcomes of refugees, research studies on immigrants have shown that among immigrants and visible minorities who have reported discrimination and/or unfair treatment, there was a greater likelihood of experiencing health decline over time [15]. In addition, mental health status may be related to the physical health status of refugees. Previous literature has reported a high prevalence of mental health concerns such as posttraumatic stress disorder, depression, and anxiety among refugees [2123]. Some research has suggested that the mental health status of refugees may also be related to the increased prevalence of physical health problems [24, 25].

Acculturation indicators may also influence the health status of both refugees and immigrants. In terms of language ability, one study reported that as residency in Canada increases to over a decade, immigrants have been found to significantly improve in their English and/or French language abilities, and they have also reported an increased sense of belonging to their community [16], which may be beneficial to their health. Furthermore, social capital involves the formation of social networks that may open up opportunities for the development of supportive relationships and also may increase access to resources through these networks [26]. Indeed, one Canadian scoping review suggested that social capital may have a positive impact on the emotional health of refugees by reducing isolation and loneliness, helping to manage stress, and building coping skills [27]. Less is known about how social capital may influence the physical health outcomes of refugees, however, since immigrants and refugees are exposed to similar postmigration experiences, such as discrimination, language acquisition, and other settlement challenges, it is of value to compare refugees to immigrants with respect to their self-reported physical health. Any differences between the two groups may be due to premigration traumas and challenges that are markedly different in the two groups.

The aims of the current study of working-age Canadians (i.e., aged 15–64) are as follows:(1)To investigate if refugees have comparable, better, or worse self-reported health than that of other immigrants and those born in Canada.(2)To identify what factors are associated with good self-reported health among refugees in Canada and to explore these relationships among immigrants and among those born in Canada.

2. Methods

2.1. Sample

Data were obtained from Cycle 27 of the General Social Survey (GSS-27), which was administered by Statistics Canada in 2013. The present study’s sample totaled 17,082 respondents aged 15–64; the sample included refugees (n = 753), immigrants (n = 5,063), and Canadian-born (n = 11,266).

2.2. Measures
2.2.1. Outcome of Interest

Self-reported health was measured through the question, “In general, would you say your health is excellent, very good, good, fair, or poor.” Self-reported health measures have a high predictive validity and are considered to accurately predict mortality [28]. Responses were dichotomized into “good, very good, or excellent” vs “fair or poor,” hereafter, referred to as “good” vs “poor.” Collapsing self-reported health into a binary variable (fair/poor vs excellent, very good, and good) has been commonly used by researchers examining self-reported health in population-based samples for the past 40 years (e.g., [2931]). Dichotomization has been identified as “a useful strategy for increasing the reliability of SRH in the general population” [31] and findings are generally consistent in analyses using the binary variable in comparison to the 5-level variable [32]. Another advantage of dichotomizing self-reported health is that it “helps account for imbalances resulting from low numbers of respondents in the extreme lower ends of the scale (i.e., those reporting poor health).” [32] In the current analysis, we were constrained by small sample sizes in the very poor SRH and refugee cells. Due to Statistics Canada’s restrictions on minimum cell size release, we were not able to report some bivariate findings and had minimal power in logistic regression analyses.

2.2.2. Key Exposure of Interest

Refugee/immigrant status was determined with the question: “Under which of the following broad immigration programs did you become a landed immigrant in Canada?” Answers were categorized into (born in Canada, immigrant, and refugee).

2.2.3. Other Variables in the Analyses

The present study included various GSS-27 questions inquiring about sociodemographic characteristics including self-reported sex (male and female), age (15–44 vs. 45–64), education (high school graduate or less and more than a high school degree), total household income (<$50,000, $50,00–99, 999, $100,00 or more, missing/refused category), marital status (single, widowed, divorced, separated vs married, and common-law), and visible minority status (white vs visible minority).

In addition, several variables related to participants' social experiences and behaviours in Canada were included. Respondents were asked if they had a confidant with the question, “How many close friends do you have (nonrelatives), who you feel at ease with, who you can talk to about what is on your mind, or who you can call for help?” and responses were dichotomized into “none” versus “one or more.” Participants were asked about their social trust and their responses were dichotomized into “hesitant to trust people” versus “most people can be trusted.” Participants were asked about their involvement in groups, organizations, or associations in the last year which was dichotomized into “none” versus “one or more.” Respondents were asked whether they had experienced discrimination, in any form, in Canada over the past five years and responses were dichotomized into “no discrimination” versus “experienced discrimination.” Participants responded in the language that they most often speak at home and responses were dichotomized into “English or French” versus “other languages.” The number of friends who spoke the same mother tongue as the respondent was also measured and responses were dichotomized into “less than half” versus “half or more.” The number of friends that the respondent had been in contact with during the last month who were from a visibly different ethnicity was measured and responses were dichotomized into “less than half” versus “half or more”.

Respondents were also asked “What is your sense of belonging to Canada?” which was dichotomized into “very strong” versus “somewhat strong, somewhat weak, very weak, and no opinion.” Self-reported mental health was dichotomized into fair/poor vs good/very good/excellent.

Two questions were only asked of newcomers (i.e., refugees and immigrants), and not of those born in Canada. These questions included 1)place of birth(grouped into America, Europe, Africa, Asia, and Oceania) and; 2) years since arrival in Canada.

Due to sample size constraints, in the logistic regression analysis that focused solely on refugees, several variables were collapsed into fewer categories including household income (<$50,000, $50,00+, missing/refused category), age (15–44, 45–64), and education (high school graduate or less and BA or college diploma or higher).

2.2.4. Statistical Analysis Plan

Bivariate analyses including the chi-square test compared Canadian-born, refugees, and immigrants for the complete sample of respondents (n = 17,082) between the ages of 15–64. Binary logistical regression analyses were used to examine the association between immigrant status and the adjusted odds ratios (AOR) of self-reported physical health when other relevant variables were simultaneously controlled. A correlation matrix was created to verify that multicollinearity was not a problem. Correlations were within acceptable levels, indicating that there were no issues with multicollinearity: Only three pairs of variables had a correlation above 0.2: marital status and age (r = 0.45), marital status and education status (r = 0.245), and visible minority status and immigrant/refugee/Canadian-born status (r = 0.596). Additional logistic regression analyses were also conducted in a subsample of respondents who were refugees (n = 753) to determine which factors were associated with the self-reported physical health of refugees. This logistic regression analysis was repeated for immigrants (n = 5,063) and Canadian-born (n = 11,266) respondents. Analyses were completed using version 26 of SPSS (IBM). Unweighted sample sizes were presented and all analyses were weighted to represent the probability of selection.

3. Results

As shown in the bivariate analyses in Table 1, refugees did not differ significantly from immigrants or from those born in Canada with respect to self-reported physical health, with more than nine out of ten members of each group reporting good health. However, refugees were significantly () more likely to be men, aged 45–64, visible minority members, with household income below $50,000, who were hesitant about trusting people, were without a confidant, and were not involved in social organizations compared to their Canadian-born or immigrant working-age counterparts. A much higher prevalence of refugees (77%) and immigrants (63%) belonged to a visible minority group compared to those born in Canada (6%) (). Approximately 40% of both refugees and immigrants reported that they had experienced discrimination in Canada in the preceding 5 years, which were significantly higher than the 32% of the Canadian-born population (). Refugees and immigrants had a higher prevalence rate than those born in Canada of speaking a language other than English or French at home, reporting few or no friends who had the same mother tongue and reporting all or more of their friends who were of a different ethnicity. The prevalence of individuals who had more than a high school degree was comparable for refugees (62%) and those born in Canada (59%), but the percentages in both these groups were much lower than immigrants (75%). Refugees arrived in Canada an average of 19.6 years ago (SD = 12.1) which was comparable to immigrants who arrived in Canada an average of 19.0 years ago (SD = 13.8). These differences were not statistically significant ().

In Table 1, a logistic regression analysis of the full sample (n = 17,082) is provided. In regard to the odds of self-reporting good health, refugees did not significantly differ from Canadian-born or immigrant respondents. Neither visible minority status nor sex was significantly associated with the odds of self-reported health.

Compared to the 45–64 age group, respondents who were 15–44 had over double the odds (aOR = 2.38) of self-reporting good health. Other statistically significant factors associated with the odds of good self-reported health include higher levels of education and higher household income.

In terms of social capital, respondents who believed most people can be trusted were 56% more likely to report good health (). Involvement in social groups, organizations, or associations in the past 12 months was associated with 38% greater odds of good self-reported health (). Those who did not experience discrimination in Canada within the past five years had 81% higher odds (aOR = 1.81) of self-reporting good health (). Those who spoke English or French at home had 21% lower odds of good self-reported health.

Table 2 presents the sample characteristics of Canadian-born, refugees, and immigrants in separate analyses. With respect to refugees (middle column), neither visible minority status, age, education level, level of social trust, and language spoken at home nor sense of belonging in Canada were significantly associated with self-reported health among refugees. The prevalence of refugees who reported good physical health was significantly higher among women, married, or common-law respondents, those who were involved in a social group, organization, or association, those who had not experienced discrimination, those who reported that fewer than half of their friends spoke the same mother tongue as the respondents, and those who had missing data on household income.

For “total household income from the past year, missing/refused/do not know,” results could not be reported as the cell size was too small. Due to small cell sizes, statistics Canada rules prohibited releasing the prevalence of having a confidant and necessitating combining those with a university degree with those of a graduate degree and combining all those with an income of $50,000 and higher. Also, it required that we combined those less than 45 with those of 45 and older.

On average, refugees in poor or fair health arrived in Canada 20.3 years ago (SD = 10.3) while refugees in good, very good, or excellent health arrived 19.5 years ago (SD = 12.3). This difference was not statistically significant. On the other hand, immigrants in poor or fair health arrived, on average, in Canada 23.7 years ago (SD = 13.9) while immigrants in good or excellent health typically arrived 18.53 years ago (SD = 13.71). This difference was statistically significant.

In Table 2, in the final column which is focused on the immigrant sample, the prevalence of good self-reported health was higher among visible minority respondents, males, younger respondents, those with higher education, married or common-law correspondents, those with more social trust, those who had did not report that they had been discriminated against in the past 5 years, those who spoke languages other than English or French, those who had about half their friends with the same mother tongue, and who had half their friends of a different ethnicity, and those with a very strong sense of belonging to Canada, with good to excellent self-reported mental health.

The Canadian-born sample is found in Table 2 in the first column. Among Canadian-born respondents, the prevalence of good self-reported health is higher among younger respondents, those with more education and higher income, married respondents, those with a confidant, those with higher social trust, those involved in social groups or organizations in the past year, those who had not experienced discrimination, those who had half or more friends of the same mother tongue, those with a very strong sense of belonging to Canada, and those with good-excellent self-reported mental health.

In the logistic regression analyses shown in Table 3, the following variables were associated with self-reported health for all three groups (Canadian-born, refugees, and immigrants): a higher level of education, not experiencing discrimination in the preceding 5 years, and good to excellent self-reported mental health.

Canadian-born respondents who spoke English or French at home had significantly lower odds of reporting good mental health. In sharp contrast, among refugees speaking English or French at home was associated with higher odds of self-reported physical health. Among Canadian-born respondents, those with half or more of their friends with the same mother tongue had higher odds of reporting good health.

Although those with younger ages and those who believed most people can be trusted had higher odds of self-reported health for all three groups, and these associations only reached statistical significance for the Canadian-born and immigrant samples. Higher income was associated with better self-reported health for Canadian-born and immigrant respondents, but not for refugees.

Married refugees had higher odds of reporting good health than nonmarried respondents, but this factor was not significant for Canadian-born and immigrant respondents.

Only immigrants reported a significant association with lower odds of self-reported good health for women, for those without a confidant, for those with fewer than half of their friends from a visibly different ethnicity different from themselves, and for those with a strong sense of belonging to Canada. These were not significant factors among refugees or for Canadian-born respondents.

Although those involved in groups or associations had higher odds of reporting good health, only the largest group, those who were Canadian-born, reached statistical significance.

4. Discussion

This population-based study’s findings indicate that there were no significant differences in self-reported physical health among working-aged refugees, immigrants, and Canadian-born individuals. More than 90% of each of the three groups reported good self-reported health. It is possible that the better-than-expected health outcomes among refugees in Canada may be partially attributable to Canada’s universal health coverage. Much of the previous research showing poorer health among refugees has been conducted in the US [7, 8, 10] where large numbers of the poor do not have access to affordable health care.

However, some previous Canadian research studies have suggested that limited primary care access among refugees [33], and discriminatory interactions with health care providers [34, 35] may negatively impact health and/or health care access. Although this study did not have information on discrimination experiences specific to the health care context, the overall prevalence of discrimination was comparable for refugees and immigrants (40%); and this was only slightly higher than for those born in Canada (32%). Furthermore, visible minority status was not associated with self-reported health in the full sample or in the refugee sample.

In both the full sample and the refugee, immigrant and Canadian-born subsamples, individuals who had not experienced discrimination in the preceding five years in Canada had close to double the odds of good health compared to their peers who had experienced discrimination. This underlines the importance of anti-discrimination strategies such as mandatory cultural competency and cultural safety training for health care professionals, and a greater understanding of the refugee experience including the pre and postmigration factors that may uniquely impact refugee health [34, 36]. Additionally, workplaces can ensure that health care workers and health information reflect language diversity in order to accommodate the language needs of refugees [36].

In keeping with the previous literature, refugees in this study were older and poorerthan immigrants and Canadian-born respondents and were disproportionatelymenand visible minority members [3]. Perhaps, because of their traumatizing pre-refugee experiences, refugees were more likely to be hesitant to trust people, to be without a confidant, and to not be involved in social organizations than their Canadian-born or immigrant working-age counterparts. When the analyses were restricted to refugees, the prevalence and odds of good self-reported health were higher among those who were more acculturated (i.e., more than half of their friends without the same mother tongue) and more socially integrated (i.e., involved in a social organization in the past year).

A prominent measure of acculturation is the ability to speak the language of the host country [37]. Our study found that refugees who spoke English or French at home had twice the odds of being in good health, although this was not a significant factor for immigrants. It is important to note different linguistic preferences among refugee families in Canada may also foster well-being. For example, one study with Syrian refugee parents and children living in Toronto found that parents preferred their native dialect to be used at home, for reasons related to the maintenance of cultural identity and religious practices [38]. However, Syrian children had a preference for speaking English at home [38]. Thus, language preferences within the home are highly individual and may be impacted by age and life stage.

Related to our findings on social integration and health among refugees, one qualitative study with Syrian refugees who resettled in Canada found that major concerns among the sample were social isolation and the desire for integration within their community and with other Canadians [39]. Within that study of Syrian refugees, seeking Canadian social support also provided these refugees with additional opportunities to learn English [39]. Thus, refugees involved in social groups, organizations, or associations may have greater opportunities to form relationships with Canadians and this can help them to diversify their social groups (e.g., in terms of their mother tongue spoken). In addition, one study with refugees and immigrants suggested that involvement in groups, associations, or organizations was beneficial in building a sense of belonging to their community, providing access to mentorship, and gaining information about opportunities in Canada [40].

Another Canadian study with refugees from Zimbabwe and Sudan found that respondents reported high rates of loneliness and that some refugees reported not making any friends in Canada despite resettling five years prior [20]. Similarly, in the current study, refugees were more likely to report not having a confidant (5.7%) as compared to immigrants (4.1%) and the Canadian-born (2.5%). In terms of social support, young refugees and immigrants suggested that having the opportunity to build social connections with those from diverse ethnocultural groups was seen as advantageous [40].

There are several limitations to this study that are related to the fact the analysis was based upon an existing dataset which was not designed with the present study in mind. First, individuals without the ability to complete the survey in English or French were not included in the study. This study, therefore, may miss the most vulnerable working-age Canadian refugees and immigrants. Refugees without English or French language skills often have a harder time obtaining employment opportunities [41] and adequate health care [34].

Another limitation of the present study is that we do not know if the refugees arrived in Canada as privately-sponsored refugees or under a government assisted program. Within the last 10 years, UNHCR [2] reported that 58% of refugees arriving in Canada were privately sponsored, and some research suggests that privately sponsored refugees have had greater access to support and social connections through their sponsors, at least during the initial settlement process [42]. Although we cannot make assumptions about the program in which refugees in our sample arrived in Canada, those who were privately sponsored may have more options for social capital, integration, and acculturation experiences [42].

Most importantly, we were restricted to the subjective assessment of health status due to the questions used in the GSS, rather than the objective measures of health determined by either a medical examination or a medical chart review. Future research studies would benefit from a more objective assessment. There has been substantial criticism of reliance on the self-reported health variable as it may vary by contextual factors, particularly among those in developing countries where the community norms of objective health standards are quite low [43]. However, within the North American context, self-reported health has been found to be a robust predictor of mortality [28]. In addition, the prevalence of good self-reported physical health in the GSS is very high (>90%) and therefore there may be limited power to obtain statistical significance between those with and without good self-reported health among refugees, for whom there is a relatively small sample (n = 753).

Unfortunately, the GSS-27 was collected almost a decade ago. We were restricted to the use of the GSS cycle 27 data from 2013 because the more recent GSS cycles did not include a question on refugee status. Hopefully, future GSS cycles and other population-based Canadian data will regularly include information on refugee status which will allow for more up-to-date assessment of the physical health of refugees in Canada.

Furthermore, we do not have data on other health behaviours (such as exercise and nutrition) and health indicators (e.g., body mass index) in this dataset. Protective health behaviours (e.g., a healthy diet) might have contributed to a high prevalence of good health reports among our refugee sample, despite other literature suggesting poorer health among refugees in comparison to those born in the host country [710]. Comparatively, in previous research studies, the prevalence of specific health concerns (e.g., heart conditions) was measured [79]. In the present study, respondents were asked about their general physical health status, however, this does not necessarily mean that respondents did not have chronic health conditions, it just may be the case that these conditions were well managed or treated.

5. Conclusions

The present study indicated that working-age refugees in Canada do not differ significantly from immigrants or the Canadian-born with respect to their self-reported health status, and that the visible minority status was not significantly associated with self-reported health status in either the general population or among refugees. Refugees are less likely to report involvement in social groups or having a confidant compared to immigrants and those born in Canada. However, refugees who are more socially integrated into Canadian society have a better physical health status. This research study suggests the need for the creation and evaluation of programs and interventions that would help reduce social isolation and can improve social connections for refugees. Targeted outreach and social care interventions for socially isolated Canadians, particularly refugees, may be beneficial for their health and well-being.

Data Availability

The General Social Survey 27 (GSS-27) dataset used to support the findings of this study is freely available from Statistics Canada. Please see https://www150.statcan.gc.ca/n1/en/catalogue/89M0033X.


This article is dedicated to the co-author, Dr. Karen Kobayashi, who passed away on May 28, 2022. She devoted her career to improving the well-being of immigrants in Canada and to mentoring the next generation of immigration scholars.

Conflicts of Interest

The authors declare no conflicts of interest.

Authors’ Contributions

A.M., U.G., and K. K., E.F.T. conceptualized the study; E.F.T. developed methodology; E.F.T. did formal analysis; A.M. and E.F.T. wrote the original draft; A.M., U.G., K.K., and E.F.T. reviewed and edited the article, E.F.T. supervised the study, E.F.T did funding acquisition. All authors have read and approved the final manuscript.


The authors would like to thank the GSS-27 for providing the data for this analysis. The study was supported by the Social Science and Humanities Research (Grant number: 435-2020-0177; PI Esme Fuller-Thomson).