Immigrants’ Length of Residence and Stalking Victimization in Canada: A Gendered Analysis

Although previous studies have explored the role of gender on stalking victimization, we know very little about how female and male immigrants are exposed to stalking victimization over time after their arrival to their host society. To address this void in the literature, we use the 2014 Canada General Social Survey to compare stalking victimization among native-born individuals, recent immigrants (those who have been in Canada for fewer than 10 years), and established immigrants (those who have been in Canada for 10 years or more) separately for women and men. Applying gender-specific complementary log-log models, we find that female (OR = 0.63, p < 0.05) and male (OR = 0.46, p < 0.01) recent immigrants are less likely to experience stalking victimization than their native-born counterparts. We also find that female established immigrants (OR = 0.65, p < 0.05) are less likely to experience stalking victimization than their native-born counterparts although there is no significance difference for male established immigrants (OR = 1.01, p > 0.05). Overall, this study points to the importance of understanding the intersection between immigrants’ length of residence and gender in the context of stalking victimization in Canada. Based on these findings, we discuss several implications for policymakers and directions for future research.


Introduction
Previous studies have provided mixed evidence on the relationship between immigrant status (i.e., foreign-born status) and crime victimization. For example, Ibrahim [1] finds in Canada that immigrants are less likely to experience violent victimization such as sexual assault, robbery, and physical assault than non-immigrants. Research also finds that immigrants are similarly exposed to hate crime and cyber-victimization in comparison to the native-born in Canada [2,3]. In the United States, it was found that immigrants are not significantly different from the native-born in terms of exposure to nonfatal personal victimization [4]. Further, McCann and Boateng [5] observe that immigrants are less likely to experience hate crime victimization than those born in the United States. Although largely limited to women as victims, research also shows that immigrants appear to be more vulnerable to intimate partner violence than native-born individuals, due to their unique challenges associated with settling and adjusting to Canada [6]. Findings on these relationships are mixed, possibly due to variations in what type of victimization are studied.
While there are many types of crime victimization, the literature pays very little attention to immigrants' experience of stalking victimization. Stalking, legally known as Sexes 2022, 3 criminal harassment, includes unwanted or repeated acts of pursuit of an individual over time that are threatening and potentially dangerous [7]. There are two main characteristics that have been identified with stalking [8,9]. For one, stalking involves an action that is committed by the stalker, which involves a range of recurrent pursuit behaviours such as spying on and communicating with the victim in person or electronically (e.g., email and texting). For another, the stalker's action should result in a feeling of fear or some other comparable emotional reaction such as anxiety for people being stalked.
There are a range of adverse effects of stalking on victims. For example, the experience of stalking is positively associated with psychological and emotional problems such as stress, depression, mood disorders, and panic attacks [10,11]. Similarly, it is found that stalking increases the likelihood of hypertension and diabetes among victims [11]. Stalking also impacts victims' sense of safety and control over their lives [12]. Several studies point out that the time and cost involved in managing and pursuing stalking cases can burden victims [13,14]. Specifically, Logan and Walker [15] find that there is a higher risk of using substances such as prescription and illegal drugs as well as alcohol to reduce stress among women who experience stalking victimization during their six-month post protective order period.
Considering these social, economic, and health consequences, it is important to understand the relationship between immigrant status and stalking victimization. In particular, we aim to unpack the heterogeneity of immigrants by exploring the intersection between immigrants' length of residence and gender in the context of stalking victimization. This approach is critical for at least three reasons. First, studies in Canada and beyond find that stalking victims are more likely to be female than male, identifying stalking victimization as gendered [11,[16][17][18]. Second, immigrants' exposure to stalking victimization may change based on their length of residence after their arrival to Canada. For example, research shows that the prevalence of risk factors associated with stalking such as tobacco and alcohol abuse tends to increase among immigrants as they stay longer in Canada [19]. Third, research shows that in close-knit communities, including those of immigrants and other minority groups, concern about their victimization becoming public may deter them from acknowledging their experience of being stalked [20]. This experience may particularly be the case for some female immigrants who may be socialized to perceive stalking as an experience to be discussed discretely and privately [21].
Despite these concerns, we know very little about immigrants' exposure to stalking victimization in Canada. To this end, we aim to understand whether there is a gendered pathway to stalking victimization among immigrants in Canada. Specifically, we compare the prevalence of stalking victimization among recent immigrants (who have been in Canada for less than 10 years), established immigrants (who have been in Canada for 10 years or more), and the native-born, separately for females and males.

Participants and Procedures
This study uses data from the 2014 Canada General Social Survey (GSS) on Victimization. The GSS is a cross-sectional voluntary survey that randomly samples noninstitutionalized Canadians aged fifteen years and older living in the ten provinces and three territories. Full-time residents of institutions are excluded. This survey is useful for the current study since it contains information on demographic indices and experience of criminal victimization including stalking victimization. The survey employs a complex, multistage sampling design, with in-depth telephone interviews. More information about sampling framework can be found on Microdata User Guide [22]. The overall response rate is 53%. Statistics Canada provides the survey weights to address the non-response bias and to represent the Canadian population. Importantly, we identify that about 25% of household income are missing in this dataset. Considering that missing cases account for more than 10% of the sample, we use the Markov Chain Monte Carlo method to address this concern. Based on Rubin's rules for scalar estimands [23], we combine ten imputed datasets and averaged them to obtain mean model parameter estimates. To this end, we include 30,185 respondents (i.e., 15,395 female and 14,790 male respondents) as part of our analytical sample for this study. As a secondary analysis of the public use microdata file of the GSS, this study does not require any ethics review.

Measures
The 2014 GSS defines stalking victimization when respondents are subject to repeated and unwanted attention that causes them to fear for their safety or the safety of someone known to them. Specifically, respondents are asked the following questions: in the past five years, has anyone (1) phoned you repeatedly or made silent or obscene phone calls?; (2) sent you unwanted messages through e-mail, text, Facebook or any other social media?; (3) sent you unwanted gifts, letters, or cards?; (4) tried to communicate with you against your will in any other way?; (5) followed you or spied on you either in person or through an electronic tracking device?; (6) waited outside your home?; (7) waited outside your place of work or school or other places you were, when they had no business being there?; (8) persistently asked you for a date and refused to take no for an answer?; (9) posted inappropriate, unwanted or personal information about you or pictures on a social media site?; (10) attempted to intimidate or threaten you by threatening or intimidating someone else?; (11) attempted to intimidate or threaten you by hurting your pet(s)?; and (12) attempted to intimidate or threaten you by damaging your property? Based on these questions, we create a variable called 'stalking victimization in the past five years' where respondents are coded as 'yes' when they answer yes to one of these questions (0 = no; 1 = yes). The independent variable is 'length of residence in Canada' where respondents are categorized into three groups based on how long they have been in Canada (0 = nativeborn; 1 = recent immigrants; 2 = established immigrants). In the Canadian literature, it is a common practice to treat immigrants who have been in Canada for less than 10 years as 'recent immigrants' and those who have been in Canada for 10 years or more as 'established immigrants' [19]. In addition, based on previous studies [11,18], we include a range of demographic and socioeconomic covariates such as gender, visible minority status, age of respondents, marital status, household income, and level of education.

Statistical Analysis
We employ regression analysis to understand the role of the independent variable (i.e., length of residence in Canada) on the dependent variable (i.e., stalking victimization). Simple logit link function assumes that the distribution between two categories is relatively symmetrical. Therefore, we rely on complementary log-log link function that is suitable when the cases are unevenly distributed [24]. There are six models in total. Considering that the experience of stalking victimization is often gendered, we calculate unadjusted and adjusted estimates separately for men (Models 1 and 2) and women (Models 3 and 4). In addition, in Models 5 and 6, we introduce the interaction term between 'length of residence in Canada' and 'gender' to identify whether the relationship between the dependent and independent variable differs statistically between men and women. Findings are reported with odds ratios (ORs). ORs larger than 1 indicate that respondents are more likely to experience stalking victimization while those smaller than 1 point to lower odds of doing so. We apply sampling weights generated by Statistics Canada to all of the analyses. Table 1 shows sample characteristics. We find that a larger proportion of women (8%) experience stalking victimization than men (5%) in the past five years. For both women and men, the largest immigrant status category is native-born (80%) followed by established (15%) and recent immigrants (5%). It is also noteworthy that more women (15%) are formerly married than men (7%) while about three fifths of women (59%) and men (63%) are currently married. In terms of household income, a higher proportion of women (20%) make $40,000 or less than men (15%). By contrast, more women (60%) achieve higher levels of education than high school in comparison with men (56%).  Table 2 shows findings from regression analysis. We find at the bivariate level (see Models 1 and 3) that male (OR = 0.52, p < 0.01) and female (OR = 0.65, p < 0.05) recent immigrants are less likely to experience stalking victimization in the past five years than their native-born counterparts. After accounting for demographic and socioeconomic factors, the results largely remain consistent, suggesting that male (OR = 0.46, p < 0.01) and female (OR = 0.63, p < 0.05) recent immigrants are less likely to experience stalking victimization in the past five years than their native-born counterparts (see Models 2 and 4). As suggested with interactive models (see Models 5 and 6), we find that there is no significant difference between male and female recent immigrants at the bivariate (OR = 1.25, p > 0.05) and multivariate level (OR = 1.20, p > 0.05). Moreover, we observe at the bivariate level (see Models 1 and 3) that female established immigrants are less likely to experience stalking victimization in the past five years than their native-born counterparts (OR = 0.50, p < 0.001) although there is no significant difference for male established immigrants (OR = 0.83, p > 0.05). Accounting for demographic and socioeconomic variables, we find that results remain generally consistent in Models 2 and 4, except that the significance for female established immigrants is largely attenuated when we control for marital status. According to interactive models (see Models 5 and 6), there is significant difference between male and recent established immigrants at the bivariate (OR = 0.60, p < 0.05) and multivariate level (OR = 0.57, p < 0.05). men and women are more likely to experience stalking victimization. We also observe that men whose household income is less than $20,000 (OR = 2.16, p < 0.05) and women whose household income is $20,000-40,000 (OR = 1.50, p < 0.05) and less than $20,000 (OR = 1.85, p < 0.01) are more likely to experience stalking victimization than their counterparts whose household income is more than $100,000. Finally, it is found that men with university degree or higher are less likely to experience stalking victimization than those without a high school diploma (OR = 0.56, p < 0.05).

Discussion
Previous studies presented mixed findings on the relationship between immigrant status and crime victimization, possibly due to variations in what types of victimization are examined. In this study, we analyze immigrants' stalking victimization, considering the dire social, economic, and health consequences of victimization. We further unpack the heterogeneity of immigrants by looking at the intersection between gender and immigrants' length of residence in Canada. This approach may be useful to identify possible behavioural, cultural, and social mechanisms in which immigrants' gendered integration into the dominant society may be a determinant of their exposure to stalking victimization over time after their arrival to Canada. Using a nationally representative survey, we compare the pattern of stalking victimization among recent immigrants, established immigrants, and the native-born (separately for females and males).
We find that female and male recent immigrants are less likely to experience stalking victimization than their native-born counterparts. This observation is largely consistent with previous research. For example, Napolitano et al. [25] find that, compared to their shorter counterparts, immigrants with a longer length of stay in Italy are more likely to experience at least one form of violence in the last 12 months. In Canada, Du Mont et al. [26] show that recent female immigrants (i.e., those who have stayed in Canada for fewer than 20 years) are less likely to experience intimate partner violence than their nativeborn counterparts. Similarly, another study observes that short-term female immigrants (i.e., those who have stayed in Canada for more than 10 years) are less likely to experience intimate partner violence than long-term female immigrants (i.e., those who have stayed in Canada for fewer than 10 years) [27]. These results point out that recent immigrants may be exposed to lower rates of victimization than their native-born counterparts.
A possible explanation for this trend may be that risk behaviours associated with criminal victimization including stalking victimization are less prevalent among recent immigrants. Specifically, although risky behaviours such as alcohol and drug consumption are considered critical determinants of criminal victimization [28], recent immigrants are less likely to engage in such behaviours than their native-born counterparts in Canada [29,30]. There are two possible reasons why recent immigrants tend to avoid these risky behaviours. For one, most immigrants need to undergo comprehensive medical screening, enabling Canada to disqualify those with severe medical conditions from migration. For another, educated and healthy people are more likely to self-select to migrate to Canada than their uneducated and unhealthy counterparts. These processes may select women and men with healthy lifestyles to migrate to Canada [31]. Consequently, both female and male recent immigrants may still endorse perceptions and behaviours that promote optimal health. This context may be helpful for them to avoid risky behaviours such as alcohol and drug consumption, reducing their exposure to stalking victimization.
Furthermore, we observe that there is a differential exposure to stalking victimization between female and male immigrants after 10 years of their arrival to Canada. For males, there is no significant difference in stalking victimization between established immigrants and the native-born. This result may be explained by previous research, which indicates that immigrants over time often adopt unhealthy behaviours such as alcohol and drug use. It is documented that some immigrants, particularly those from non-European regions, tend to adopt liberal drinking norms of the host society, which can increase immigrants' prevalence of alcohol consumption over time [32,33]. Other studies also point out that some immigrants may be vulnerable to resettlement stress through unique experiences in the host society. These experiences may include racial discrimination, social isolation, and economic deprivation, all of which can contribute to their exposure to risks associated with stalking victimization such as alcohol and drug consumption [26,27]. Based on these considerations, we argue that there may be behavioural and structural conditions in which male immigrants are increasingly exposed to risk factors that are associated with stalking victimization over the course of their integration and settlement into the dominant society.
For females, however, established immigrants are less likely to experience stalking victimization than the native-born. We provide several possible mechanisms to explain this result. Research shows that male immigrants are often expected to fulfill their responsibility as global breadwinners by financially supporting both immediate families in host countries and extended families in home countries [34]. In this context, integrating into the mainstream labour market may be an important transition for male immigrants to fulfill their responsibility among male immigrants [35]. Considering that immigrants often experience unique labour market obstacles such as racial discrimination and non-/under-recognition of foreign human capital and credentials [36], male immigrants may be particularly exposed to resettlement stress as a result of pressure to be economically successful. In contrast, female immigrants often migrate to Canada through spousal reunification class after their male partners have settled and achieved some level of economic stability and viability [37]. In this case, female immigrants are not only exonerated as global breadwinners, but they may also be shielded from structural and economic stresses in the host society. Based on this advantage, female immigrants may be more protected from risky behaviours including alcohol and drug consumptions as coping mechanisms, potentially reducing their exposure to stalking victimization. Moreover, as a consequence of labour force participation through economic integration, male immigrants may be establishing diverse networks with the native-born in the dominant society [38]. Therefore, female immigrants may have fewer opportunities to interact with the native-born, possibly preventing them from modifying their lifestyles through social learning and adaptation in the host society. Based on these results, we suggest that there may be gendered integration pathways between female and male immigrants, which can facilitate female recent and established immigrants to be exposed to lower rates of stalking victimization than their native-born counterparts.
In addition to risk-related arguments, changing perceptions and interpretations of stalking victimisation can explain our results. Research suggests that some immigrants, particularly those from non-Western societies, can have very different beliefs and perceptions about criminal victimization [39]. In this context, spending time in the host society and learning social and cultural norms to construct new definitions and meanings of stalking victimization may be critical among recent immigrants [40]. While we find that both female and male recent immigrants are less likely to experience stalking victimization than their native-born counterparts, this result may be demonstrating that recent immigrants are not perceiving their experiences of stalking victimization in the way it is defined and communicated within the sociolegal context of Canada. This may be due to potential cultural differences in the construction of stalking victimization in their origin countries and host societies in Canada. Male established immigrants experience a similar level of stalking victimization to the native-born, which may be reflective of acquiring new definitions of stalking victimization through socialization in the host society although female immigrants are still less likely to experience stalking victimization than the native-born even after 10 years of their arrival. Importantly, research shows that there are social and cultural norms among some immigrant communities that the experience of gendered violence such as stalking violence should be considered private [20,21]. Due to this context, some female immigrants, irrespective of their time spent in Canada, may be inclined to accept and hold to such patriarchal values and norms, potentially making it difficult for them to establish new definitions of stalking victimization that are consistent with those in the dominant society.
Although not the focus of this article, we also find that a range of control variables are significantly associated with stalking victimization. For example, younger women and men are more likely to experience stalking victimization than their older counterparts. This finding is consistent with previous studies [41,42], which suggest that older age is often a protective factor against many types of victimization among women and men. In addition, formerly married and never married women and men are more likely to experience stalking victimization than their currently married counterparts. This result may be explained by Cao et al. [43], who identify marriage as a useful social bond that can promote a supportive environment that reduces married people's chances of stalking victimization compared to non-married people. Moreover, women and men with lower income are more likely to experience stalking victimization than those with higher income. Supporting this observation, previous research in Canada and the United States reveals that income is negatively associated with criminal victimization including stalking victimization [44,45], which may point to the possibility that financial resources are protective against stalking victimization.

Limitations
This study has some limitations. For example, it is important to capture the cohort effect, in addition to the duration effect, in our analysis. It is argued that, due to the changes in immigration policy over time, immigrants in earlier cohorts have different social, cultural, and economic characteristics than those in later cohorts [46,47]. However, the cross-sectional nature of the GSS does not allow us to explore the role of this cohort effect on stalking victimization. In addition, admission class may be a useful factor to explore the heterogeneous nature of immigrants in Canada. Future research should explore whether there are any variations in stalking victimization among economic immigrants, family-class immigrants, and refugees. Finally, as discussed earlier, we are not able to capture any cultural aspect of stalking victimization in the GSS, and this is likely to be a limitation since immigrants often endorse diverse social and cultural norms in their construct and definition of stalking victimization [48]. These limitations point to the importance of developing and implementing a comprehensive longitudinal survey that captures immigrants' experiences in stalking victimization in Canada.

Conclusions
Despite some limitations noted above, there are several implications for policymakers. For example, we find that female and male immigrants are differently exposed to stalking victimization than their native-born counterparts. In this context, it may be important to establish prevention strategies while taking the intersectionality between immigrant status and gender into consideration. Specifically, we speculate that there is a gendered pathway of immigrant integration in Canada, possibly impacting women and men differently in their adoption of risk behaviours associated with stalking victimization. Addressing this issue, it would be useful to tackle structural issues related to resettlement stress such as racial discrimination, lack of social support, and non-/under-recognition of foreign human capital and credentials. In addition, it is equally important to establish culturally and socially sensitive programs for female and male immigrants to understand the sociolegal context of stalking victimization in Canada. This may especially be the case, considering that the definition of stalking victimization is often constructed and addressed differently between immigrants' home countries and Canada. Finally, given the sensitive and gendered nature of stalking victimization, some immigrant communities may still prescribe and endorse patriarchal norms in addressing this issue. Therefore, there is an urgent need for policy stakeholders to implement community-level programs for female and male immigrants to discuss and increase awareness about gender-specific violence including stalking victimization in Canada. Institutional Review Board Statement: Not applicable since we used secondary data.