THE RECENT GROWTH OF REFUGEE POPULATIONS across Canada has prompted considerable research concerning successful resettlement. One of the first steps toward this end is for newcomers to establish stable and meaningful employment. As a refugee status implies, however, many refugees enter Canada with limited opportunities to reproduce a lifestyle similar to or better than one that was experienced in their former country. Problems with foreign credential recognition, or lost documentation verifying the occupational and educational status of refugees are common dilemmas encountered in the employment adjustment process. To compensate for this, refugees no doubt turn to their immediate social networks, or what has been referred to as social capital (Loury, 1977; Bourdieu, 1985; Coleman, 1990; Portes, 1998), to aid them. In particular, family networks, service providers, sponsors and ethnic-group members are the first sources of aid refugees seek out when they attempt to find employment.
Taking into account the value of their former employment and education, current English-language proficiency, and the number of months of training/education received upon arrival in Canada (or potential human capital), (1) this paper examines the relative worth of network ties in achieving successful employment outcomes. As an interpretive framework, Giddens' structuration theory is used to understand how these potential forms of capital function in an employment context. From this perspective, potential human capital and social ties are viewed as ever-evolving structures that may enable or constrain refugees in the resettlement process. Given that their human capital may be ineffective or constrain employment success, can refugees' social networks compensate for this potential loss? Or, conversely, would certain social networks also constrain or limit refugees in achieving successful employment outcomes?
Beyond the impacts of capital, this study recognizes that a refugee's capital power is significantly dependent upon a range of additional factors, such as gender, age, region of origin, length of residence in Canada, foreign credential recognition, and/or experiences of discrimination. Thus, employing multiple regression analysis as its main strategy, this study controls for a range of potential predictors of employment quality to isolate the relative impacts of potential forms of human and social capital.
Review of the Literature
Research has identified a range of factors determining the employment success of newcomers to Canada. Knowledge of the English and French languages, recognition of foreign credentials, level and type of education, work experience, length of residence in the receiving society, age and sex are among the key predictors of employment status, career development and income level (Hem, 1993; Haines, 1996; Pendakur and Pendakur 1997). Furthermore, Pendakur and Pendakur's (1997) study of Canadian immigrants reveals that knowledge of both official languages (English and French) acts as a valuable form of human capital. That is, having knowledge of both English and French produces better labour market outcomes than being proficient in English or French only.
Piche, Renaud and Gingras (1997) offer several noteworthy findings about employment adjustment among newcomers in Quebec. Factors such as age, sex, education and work experience were found to significantly influence income levels and socio-economic status. In particular, they found that, in addition to work experience, being young and male facilitates finding a first job (Piche et al., 1997). In addition, as duration of residence increases, Piche et al. (1997) suggest that immigrants develop strategies, such as establishing effective network structures, to improve their employment opportunities.
Several factors that may prevent newcomers from achieving their employment goals are also part of the adjustment process. For example, a recent study released by the Canadian Race Relations Foundation (Kunz, Milan and Schetagne, 2001) supports the position that subtle forms of discrimination, such as not being considered for a promotion or being excluded from the "inner circle" at their place of employment, can seriously undermine career advancement. This is particularly the case for visible minority immigrants (Kunz et al., 2001).
To complicate matters further, various forms of systemic discrimination may be operating to effectively deny refugees access to professional-level occupations. In order to maintain high standards, professional associations (for example, in medicine) and trade unions function as labour market shelters (Krahn and Lowe, 1998) by applying stringent standards in the adjudication of foreign credentials. Krahn et al. (2000) also draw attention to the increasing institutionalization of downward mobility exercised by educational institutions. Edmonton's Grant McEwan Community College, for instance, offers a "fast-track" nursing program for former medical professionals such as cardiac surgeons, gynecologists and radiologists (Ohler, 1999). Institutionalizing the rejection of foreign credentials reinforces the barriers that prevent access to employment commensurate with previous education and experience.
To compensate for these barriers, research has shown that newcomers turn to kin and friends to increase their employment opportunities (Hem, 1993; Gold and Kibria, 1993). Despite kin and co-ethnic support networks, however, newcomers may remain at a disadvantage, particularly if they lack employable skills. Upward mobility may be less likely for those who remain within co-ethnic group businesses. The search for economic stability is particularly problematic for newcomers who belong to ethnic groups that have limited resources. Although co-ethnic networks may be used as "strategies for survival" (Gold and Kibria, 1993), they may function little beyond that. For example, Gold and Kibria (1993) found that within a co-ethnic structure, aspiring entrepreneurs among Vietnamese refugees were unable to access necessary forms of capital, such as large sums of money, as well as knowledge about licensing, competitive store locations and so on. Similarly, Pendakur and Pendakur (1998) found that ethnic enclaves in Canada primarily serve immigrants with lower levels of schooling and skills. In contrast, they did not find evidence of ethnic enclaves that have industries to support immigrants with higher levels of schooling.
This brief review reveals a critical gap in the current literature on refugee resettlement. That is, while many researchers acknowledge the influential role of network formation in refugee resettlement, most research does not specify the constraining or enabling impacts of these ties in the resettlement process. In particular an examination of refugees' labour force participation within contexts of constraining and enabling network structures has received scant attention in the literature. Moreover, a critical gap in the literature exists with respect to how refugees negotiate their resettlement with potential forms of human capital within the context of networks and larger societal forces.
Giddens' Structuration Theory Capital and Employment Outcomes for Refugees
Anthony Giddens' (1984) structuration theory examines the relationship between agency and structure. Arguing that agents have knowledge of society's institutions and an understanding of the consequences of their actions, Giddens stresses that actors are continually making choices. These choices, however, are made within the boundaries of an evolving structural context. Structure entails two sets of properties: rules and resources. In their active negotiation with society, individuals draw upon these rules and resources to affect intended outcomes. Put another way, all action is situated within these intersecting sets of rules and resources.
Describing the duality of structure, Giddens (1979: 69) views rules and resources as both enabling or constraining, acting as "both the medium and outcome of the practices that constitute [social systems]." Rules offer prescriptions for behaviour, such as those embedded in a gender script. If violated, these rules can give rise to negative sanctions. Thus, rules generate practices, acting as the medium in the production and reproduction of social systems.
Giddens (1979: 69) treats resources as the "bases" or "vehicles of power," offering individuals "facility" to create options or choices when acting. In other words, "resources are the media through which power is exercised" (1979: 91). Furthermore, power is not a resource itself, rather it is manifest or instantiated in action and remains latent if not used. In the case of refugees, capital can be defined as resource; capital potential represents the latent power inherent in the capital; and capital power represents an actor's capability to produce intended actions and outcomes. Capital as resource can be argued to offer refugees transformative capacity, whereby the power inherent in the resource enables desired resettlement outcomes.
Giddens argues further that in all social systems there are "shifting balances of resources, altering the overall distribution of resources." However, Giddens (1982: 197-98) also argues that no matter how unbalanced the distribution of power, actors who are seemingly powerless, such as refugees, "are able to mobilize resources whereby they carve out 'spaces of control' in respect of their day-to-day lives and in respect of the activities of the more powerful."
At several critical levels in her or his resettlement, a refugee's adjustment process can be interpreted within this framework of structure, agency and power. By defining structure as both enabling and constraining, we can view refugees as active agents in the resettlement process. How resources are mobilized, power is enacted, and outcomes produced depend upon the rules governing social practices. For example, in the case of their employment quality, a refugee's employment choices can be argued to depend upon capital power. Agency, in this case, is implied in the development of social networks and how refugees use network ties to find employment. However, refugee actions are determined by the responsibilities and rules inherent to the social tie developed. For example, family ties may constrain a refugee, depending on the rules governing child and kin care practices. Actions are also determined by refugees' human capital resources, including former education and work experience, as well as their English-lang uage ability Furthermore, external structural barriers, such as the rules governing foreign credential recognition procedures, may determine how these resources are used.
Data Collection and Participant Characteristics
This adult refugee sample is drawn from a larger study which involved refugees destined to Alberta between 1992 and 1997 (Abu-Laban, Derwing, Krahn, Mulder and Wilkinson, 1999). This multi-phase study examined the economic and social integration experiences of 525 refugee adults and 91 youth. Records from Citizenship and Immigration Canada (CIC) show that 9,198 refugees were destined to Alberta between 1992 and 1997. For the purposes of the study, CIC provided contact information for 5,208 of these refugees. The sample contained refugees who were both government-sponsored and privately sponsored. Dependents and refugee claimants were omitted from the sample. A systematic sampling strategy (every Nth name or family unit) was employed to arrive at a target sample of 956 individuals. Out of this target sample, only 47 (5%) individuals could not be located. Because the remaining 909 individuals were living in communities across Canada, it was not feasible to interview all of them.
In total, 616 refugees, destined mostly to Alberta, were interviewed (525 adults and 91 youth). Seventy-four interviews were conducted with refugees who, destined to Alberta originally, were now living outside of the province. Out-of-province interviews took place in British Columbia (29), Saskatchewan (1), Ontario (41), Quebec (1) and Nova Scotia (2). The remaining interviews (451) were conducted in cities across Alberta, including Edmonton, Calgary, Lethbridge, Red Deer, Medicine Hat, Grand Prairie and Fort McMurray. The current study focusses exclusively on the employment experiences of the adult sample population. While potentially employable, youth are not considered in the analysis, as many were in school full-time when they were interviewed.
Slightly more females (50.5%) than males (49.5%) are represented in the total adult sample. Seventy-three percent of respondents are married, 70% of whom are living with their spouse. The remaining respondents are single (17%), separated/divorced (4%), widowed (3%), or in a common-law relationship (3%). Almost one half of the adult refugee sample falls between the ages of 31 and 40 years.
Respondents from a range of source areas were interviewed, including refugees from Africa, Central/South America, East Asia, the former Yugoslavia, the Middle East and Poland. A significant majority of respondents (61%) were from the former Yugoslavia, even though Yugoslavian refugees made up slightly less than half of the Citizenship and Immigration Canada database from which the sample was drawn. This over-sampling is mainly a result of the more recent arrival in Alberta of refugees from the former Yugoslavia. As a result, they were somewhat less likely than other members of the 1992-1997 refugee cohort to have moved out of Alberta. In turn, they were somewhat easier to contact and, ultimately, more likely to be interviewed (Abu-Laban et al., 1999).
A structured interviewing format was developed, which included a significant proportion of fixed-response questions and a number of open-ended questions designed to represent the range of individual opinions and experiences. Interviews were conducted between July and October 1998. Thirteen interviewers, chosen for their understanding of refugee issues and multilingnal and interview skills, conducted the interviews. Face-to-face interviews were completed for most of the refugees living in Alberta. About two thirds of out-of-province interviews were conducted by telephone. Interviews were completed in a range of languages representing the various linguistic groups, with approximately one-third of all interviews conducted in English.
To determine the impacts of human and social capital on the quality of employment, a single index was developed for refugees' quality of employment, the dependent variable in the study Additional indices were constructed for each set of independent variables--potential human and social capital. Discussed below are refugees' current employment status, their current stock of capital, and how employment quality and potential forms of capital are operationalized for the multiple regression analysis.
Quality of Employment
A large majority (82%) of adult refugees held a paying job in Canada at some point after arrival. More males than females were employed following arrival in Canada (89% versus 75%, respectively), while rates of Canadian employment increased with age. In general, quality of employment within this sample is low. At the time of being interviewed, 70% of the sample were not satisfied with their current occupation, while 60% reported being over-qualified for their current job. Refugees trained in managerial! professional careers suffered the greatest downward mobility While 39% of the total sample were employed in professional or managerial careers in their former homes, only 7% were able to find comparable employment when they first arrived in Canada (Krahn, Derwing, Mulder and Wilkinson, 2000). When interviewed, over three quarters of the refugees who were trained in managerial and professional careers remained in jobs below their educational training level. Furthermore, a large proportion of the refugees (75%) had not received a promotion in their current job.
Steady employment was not common. The unemployment rate at the time of the interviews was 16%, more than double that of the annual average unemployment rate (5.7%) for Alberta in 1998 (Statistics Canada, 2000). Non-standard employment also added to the generally low quality of employment, with 30% of all refugees employed in temporary jobs and 28% in part-time jobs. Women (66%) were more likely than men (34%) to be in part-time employment, and had higher rates of unemployment (19% versus 14%).
The dependent variable in this study, a cumulative index measuring quality of employment, was constructed to include six main dimensions: current occupational status, full-time/part-time status (one part-time or several part-time jobs), temporary (has a specific end date) or permanent status, promotion status, over-qualification status and employment satisfaction. High values on this index indicate a higher quality of employment (see Appendix Al). Twenty-six percent of refugees scored from 0 to 1, 46% from 2 to 4, and 7% had scores of 5 or 6 on the quality of employment index.
To measure refugees' human capital upon arrival in Canada, four predictor variables were considered: a refugee's former occupation; their former education; the number of months of English-language training received in Canada, and the number of months of training/education obtained in Canada (see Appendix Al). A profile of a refugee's stock of human capital reveals that about one third of the sample were employed in professional/managerial positions, while the remaining two-thirds were employed in semi-professional or blue-collar jobs in their home country. A vast majority (81%) had received a high-school diploma, with over one half having received some form of post-secondary education. When they arrived in Canada, most refugees were enrolled in some form of English-language training, the majority (71%) receiving up to one year of language training. Over one third (37%) of the sample obtained some other form of additional training or education in Canada. Twenty-nine percent of the sample received up to one yea r of additional training. Only a small minority (6%) of respondents received between 14 and 24 months of training, and even fewer (2%) spent more than two years obtaining additional training.
Several indicators used to determine refugees' network structures were combined to form two indices of social capital: proximity to close family members; and the presence of extra-familial ties. In addition, the network ties responsible for helping refugees find a job were also considered (see Appendix A2).
An examination of their network structure shows that the vast majority of refugees arrived with some family members when they entered Canada (86%), a significant proportion coming with a spouse and one or two children. Upon arrival, the majority of refugees (over 70%) in the sample were married and living in a household with 2-4 members. Furthermore, slightly over one quarter of the refugee sample had family members living in the same city as the refugee. In particular, siblings and adult children were the main family ties directly surrounding a refugee's family household structure.
Refugees in the sample were connected to a range of extra-familial ties, such as friends, neighbours, other newcomers, people from work and sponsors/host volunteers. A large proportion of the sample reported seeing friends from their own culture "daily" or "often" (47%). Refugees spent the least time with neighbours, with almost one half (47%) reporting spending no time with this network tie. Over one half (52%) of the total sample visited with other immigrants "sometimes." Similarly, almost one half (49%) of all refugees reported visiting other Canadian friends "sometimes." Refugees also spent time with co-workers outside their place of work, with over one third reporting visiting with co-workers "sometimes" and another 14% visiting with them "often." Seventeen percent of refugees maintained contact with their sponsors, while 19% of the respondents still spent time with a host volunteer.
A final measurement of social capital involves identifying the network ties that were responsible for helping refugees find a job (see Appendix A2). Refugees were asked to respond to the following question: "How did you get this (current) job?" One quarter of the sample received help from in-group ties such as friends, family and ethnic-group members to find employment. Refugees relied significantly less on ties outside their in-group; only 3% relied on their sponsor or former employer to find a job, while a similarly small proportion (5%) were successful in finding employment when enlisting the aid of government and social service agencies.
Recognizing that employment quality is influenced by factors other than a refugee's stock of capital, the analysis controls for a number of additional factors. Based on previous literature suggesting their potential impacts, variables controlled for in the analysis include ascribed characteristics such as sex, age, and visible minority status. Sex and age distributions were discussed above. Males were given a score of 1, while females were assigned a score of zero. Age scores were calculated by year of birth. A refugee's visible minority status was determined by categorizing former Yugoslavian and Polish immigrants as non-visible minorities and Middle Eastern, East Asian, African, and Central/South American refugees as visible minorities. Over one third of the sample included members of visible minority groups.
Because of the possible trauma faced in their country of origin, refugees' health status may play a critical role in securing employment. Respondents were asked to judge their physical and psychological health according to a four-point scale ("very healthy," "somewhat healthy," "unhealthy," very unhealthy"). Both psychological and physical health scores were averaged to obtain an overall health score. Almost two thirds of the total sample reported feeling "very healthy" physically upon arrival, while just over half (57%) of the refugee sample stated that they were "very healthy" psychologically upon arrival.
Understanding that employment opportunities may also be determined by place of residence, the size of the refugee's current city is considered, the assumption being that larger centres may offer better and more employment opportunities than smaller cities. Cities were collapsed into three categories, where "one" represents a small city, "two" a mid-sized city, and "three" a large city, such as Edmonton or Calgary. Almost 60% of the sample lived in large cities. Mobility factors, such as a refugee's departure from his or her destined community and plans to move from the current city of residence are also considered, since refugees may believe that better employment opportunities can be found in other locations. Binary variables were constructed for both mobility factors. Over three quarters of the sample stayed in the city to which they were destined. Under one half of the sample (44%) reported that they were planning to move from their current city of residence.
Because they have been identified in research as important factors in employment adjustment, length of time in Canada and length of time in current city of residence are also controlled. Length of time in Canada was measured in years. Two thirds of the total sample arrived between 1994 and 1996. Length of time in current city of residence was measured in months. Just under one half had been living in their current city of residence between 12 and 36 months, and one third between 36 and 60 months.
Employment barriers, such as discrimination and problems with foreign credential recognition, are also controlled. Binary variables were constructed for both of these factors. One quarter of the sample had experienced some form of discrimination. A large proportion (55%) of those who did report being discriminated against experienced employment discrimination specifically. Problems with foreign credential recognition may also reduce a refugee's quality of employment. A large proportion, about 42%, had experienced problems with foreign credential recognition. Time spent in a refugee camp is another possible deterrent, suggesting that a refugee was unable to work or perhaps indicating that a refugee's education was interrupted. Time spent was measured in months. Approximately one third of all respondents reported spending some time in a refugee camp. Ten percent of the sample spent up to 12 months in a camp, 15% spent between one and five years in a camp, and a small proportion of refugees (6%) spent more than five years in a camp.
Multiple Regression Results: The Impacts of Human and Social Capital on Refugees' Quality of Employment
The Impact of Human Capital Traditional applications of human capital theory would single out one's occupation and educational attainment as having the greatest impact on employment success (Krahn and Lowe, 1998). However, when considering a refugee's formerly obtained education (beta = .003) and former occupational status (beta = .002), these two variables account for virtually none of the variation in quality of employment ([R.sup.2] = .001).
Furthermore, in applying human capital theory to the case of refugees, we should expect that a higher advanced education, a high-status occupation in one's former home, fewer months of English-language training and additional training obtained in Canada should result in a higher quality of employment. However, when considered collectively, these four human capital variables predict a very small proportion of the total variance in refugees' quality of employment ([R.sup.2] = .045). These findings suggest that refugees' performance in the labour market is largely not determined by their prior education or work experience, or by additional human capital acquired in Canada.
How does one's stock of social and additional control variables impact upon a refugee's quality of employment? Table 2 shows that when considering the combined impact of all forms of capital and control variables on quality of employment, we can account for approximately 22% of the variance in the dependent variable.
In terms of human capital, the impact of former occupation and education remains non-significant when entered into the larger regression equation. Table 2 reveals that a refugee's former education (beta = -.05) and occupation (beta = .08) play inconsequential roles for a refugee in the Canadian labour market. Consistent with these findings, it is also true that as refugees experience greater difficulty in having their foreign credentials recognized, their quality of employment decreases significantly (beta = -.10*). Thus, gaining recognition for foreign credentials is a major barrier for refugees in their attempt to secure employment comparable to their former careers.
Moreover, the problem of foreign accreditation may not be rectified by additional training or education in Canada. Surprisingly, results reveal that as refugees obtain more training/education, their quality of employment decreases (beta = -.10*). At first thought, this seems unlikely. However, this finding may simply reflect the impact of time invested in education/training.
While a longer time invested in obtaining additional training may reflect a need to improve upon current skills, and hence act as a barrier to achieving a higher quality of employment, it may also be true that as refugees invest time in training, the process of accessing higher-status employment opportunities is postponed and worthy opportunities may be missed. This time-investment factor may also be applied to the months invested in English-language training. Table 2 shows that as refugees require more months of English-language training, their quality of employment drops significantly (beta = -.20**). In addition to possibly reflecting a lower proficiency in English and hence, acting as a barrier to finding high-quality employment, it is also likely that the more months a refugee needs to invest in English-language training, the less time he/she has to be exposed to networks that may open the doors to advanced career opportunities.
Experiences of discrimination also interfere with positive employment outcomes (beta = -.1l*). Visible minority refugees (24%) in the sample are more likely than other refugees (9%) to report instances of employment discrimination, suggesting that racial discrimination may play a role in negatively affecting employment outcomes. Subtle forms of discrimination and systemic barriers, such as labour market shelters and institutionalized downward mobility, may be at play as well.
Up to this point, the regression analysis highlights two critical findings about human capital. First, one's former education and employment have little, if any, impact in securing meaningful and stable employment in Canada. This outcome is particularly true for refugees who had high-status positions (i.e., professional/managerial) in their former country. The former education and employment of these refugees is virtually worthless; at most, it qualifies them to pursue additional education in Canada that, as the results suggest, does not guarantee them entrance into their chosen fields. Second, structural barriers, such as discrimination and time invested in English-language and other training, may thwart or postpone positive employment outcomes. These findings lead to the conclusion that refugees' own personal resources or human capital appear to be either lacking or insufficient (due to structural barriers) as they struggle to gain employment comparable to what they had in their former homes. But what happe ns to refugees' quality of employment when they reach beyond their own stock of human capital and receive aid from the network ties surrounding them? Can refugees' network ties compensate for the lack of capital power inherent in their human capital?
The Impact of Network Ties Of all the forms of capital considered, a refugee's network structure shows the greatest impact on quality of employment. The largest positive impact comes from living with a spouse/partner. Table 2 shows that refugees who live with their spouses or partners are more likely than those who are not to have a higher quality of employment (beta = [.20.sup.**]). (2) In this case, it is possible that living with a spouse or partner per se does not improve one's quality of employment, but instead reflects the educational and occupational status of individuals in a married/cohabiting unit. That is, perhaps refugees with a high quality of employment are more likely to be married or living with someone with a similar or equivalent occupational status.
However, another explanation is possible. Perhaps the answer lies in examining not what refugees have but what they lack. Without a working spouse in the household, refugees may not have the flexibility to search for high-quality employment opportunities, being forced instead to accept the first job that is available. That is, living with a spouse may offer more financial support, the time to pursue a variety of employment options, and may expose refugees to their spouses' network ties, some of whom may help refugees in finding employment. It may also be the case that the single refugees in the sample may be in the process of pursuing an advanced education and are supporting their schooling with part-time or full-time, lower-status positions.
However, whereas living with a spouse may offer opportunities to enhance one's quality of employment, the presence of other family ties may produce the opposite effect. Table 2 shows that living close to parents, siblings or adult children and the presence of children in the household tends to lower a refugee's quality of employment (beta = [-.12.sup.*]). Living close to family members may constrain refugees from achieving a high quality of employment because of child care or extended kin care (i.e., parents or, perhaps, siblings) responsibilities. This may be the case for women, in particular. (3)
The positive impact of a refugee's network structure is also strongly evident when controlling for contacts that are self-reported by refugees as having a direct influence in helping them find a job. That is, refugees who used some form of social capital, particularly friends and family, to find a job were more likely to have a greater quality of employment (beta = .17 **) than those individuals who relied strictly on their own human capital (work experience, education and training) and/or their own personal efforts (applying for jobs, dropping off resumes, consulting the Internet/yellow pages, etc.) to secure employment.
Despite this significant impact, however, one cannot ignore the evidence that a refugee's quality of employment in Canada is still below the quality enjoyed in his or her former home. Refugees largely remain underemployed in Canada relative to their previous employment status, this particularly being the case for those refugees whose prior occupations were in the professional/managerial sector. As mentioned above, at the time of being interviewed, over three quarters of the refugees who had professional/managerial careers in their former home remained in jobs below this occupational level in Canada. Thus, network ties, particularly in-group ties, may be helpful in locating employment for refugees, but they cannot restore refugees' previous occupational status. Thus, not only is a refugee's human capital upon arrival undervalued and underused, but the networks they are presently employing may not be sufficient to com ensate for their downward occupational mobility
Control Variables As predicted in previous research, gender, age and visible minority status also contribute to a refugee's quality of employment. However, they do so to a lesser degree than do particular forms of capital. Regarding gender differences, refugee men are more likely than refugee women to have a higher quality of employment (beta = .10 *), reflecting the broader literature on gendered inequalities in the workplace (Krahn and Lowe, 1998). The effect of age is significantly more apparent; younger refugees experience a significantly higher quality of employment than older ones (beta = -.15 *). Because refugees' former occupations are devalued in Canada, older refugees cannot benefit from their accumulated years of work experience, a human capital advantage that other Canadians use to advance in their careers. Indeed, older refugees were more likely than their younger counterparts to claim that they were overqualified for their positions. In addition, none were represented in professional/managerial careers, proportionately fewer benefited from promotions, and fewer had full-time employment. A refugee's visible minority status also strongly impacts employment quality; visible minority refugees are significantly more likely to face a lower quality of employment (beta = -.14 *).
Beyond these ascribed characteristics, one additional control variable has a significant impact on quality of employment. That is, the longer a refugee has lived in his or her current city of residence, the higher his or her quality of employment (beta = .21 *). This finding suggests that residential stability may increase opportunities to broaden the range of network ties instrumental in employment adjustment and advancement.
Traditional applications of human capital theory that identify the value of occupation and education as a major determinant of subsequent employment success are challenged in this study. Giddens' structuration theory is useful in understanding a refugee's employment success as it highlights four elements central to the refugee resettlement process: structure, capital, power and agency Specifically, structuration theory draws attention to the rules and resources governing refugees' resettlement outcomes. In terms of human capital potential, forms of discrimination and the rules governing foreign credential recognition procedures may place limits on individual agency as refugees attempt to pursue high-quality employment in the face of these larger structural barriers. In the context of new societal rules, the transformative capacity formerly inherent in refugees' human capital was disabled. While focussing primarily on refugees from Alberta, these results are generalizeable to the country as a whole. While it c ould be argued that a stronger economy may exist in Alberta compared to other provinces, the results of this study suggest that restrictions placed on human capital use would be even more pronounced for refugees in other parts of the country.
Furthermore, although needed, the time invested in English-language and other training postpones further the chance to pursue any career advancement opportunities. To compensate for this loss in power, refugees had no choice but to re-examine the value of their human capital. Supporting Giddens' claim that actors have knowledge of societal institutions and the consequences of their actions, refugees in this sample were acutely aware of their deficiencies and focussed on re-establishing productive resources. In fact, it could be argued that it is those individuals confronted with and challenged by a new set of structural rules governing social practices who are more likely to be mindful of their subordinated position in society, and who are also more likely to be aware that they need to "know how to play according to the [rules]" (Giddens, 1979: 67).
To compensate for the low value of their human capital, refugees also turned to their close ties. In the context of structuration theory, network ties are structures that can be understood as both enabling and constraining. Regarding family ties, living with a spouse or a partner may, in Giddens' terms, enable refugees greater flexibility in pursuing educational and employment opportunities. Furthermore, refugees who chose to enlist the aid of familiar and trusted in-group ties, such as family and friends, received a higher quality of employment than those who relied strictly on their own human capital or personal efforts. Thus, despite the low value of their human capital, refugees turned to their limited pool of ties, and proved to be "very adept at converting whatever resources they [did] possess into some degree of control" over their economic circumstances (Giddens, 1982: 199).
However, what this study reveals, more importantly, is the limited value of certain network ties. When considering the quality of a refugee's current employment, the results suggest that while family and ethnic-group ties can aid refugees in their employment adjustment, these ties may not be able to compensate for refugees' downward occupational mobility. This is particularly the case for refugees who were previously employed in professional or managerial positions. Ethnic-group and family ties do not have the capital power to overcome societal restrictions, such as regulations regarding foreign credential recognition or institutionalized downward mobility. Functioning mainly as "strategies for survival," ethnic-group networks are limited in their use, particularly if these in-group ties have inadequate access to resettlement services and limited knowledge about career development and advancement opportunities. Furthermore, this study cautions that the miles governing tie relations may actually constrain a re fugee's quality of employment. Refugees (perhaps, especially, females) may be at a social capital disadvantage, as proximity to close family members may be indicative of family care responsibilities, thereby limiting employment advancement. Thus, as forces of constraint, certain network ties engender rules that involve the reproduction or negotiation of roles, such as primary caregiver. Thus, rules may have the power to constrain refugees from producing certain outcomes (i.e., pursuing employment opportunities), but may also act (are enacted) to ensure the reproduction of others.
Understanding how refugees integrate into the Canadian labour force requires that policy and program makers look to the type of support refugees are drawing upon when they first arrive in Canada. Taking stock of the types of ties that refugees draw upon would allow service providers to integrate these sources of aid into the resettlement process and to focus on improving their effectiveness or reducing their constraining impacts. Toward strengthening a refugee's social capital power, service providers can supply network ties with appropriate and useful knowledge about the Canadian labour market and its structural restrictions, as well as opportunities to build an extensive range of ethnic group-based resources ranging from help in caring for dependents to opportunities for career advancement.
Quality of Employment and Human Capital Measurements
Quality of Employment Index
Professional/Managerial (Binary Variable)
Full-time/Part-time Status (Binary Variable)
Permanent/Temporary Status (Binary Variable)
Received Promotion(s) (Binary Variable)
Overqualified for Occupation (Binary Variable)
Respondents scored one point each if they had a professional or managerial position, if they were in a full-time, permanent job, if they received a promotion, and if they believed they were not overqualified for and were satisfied with their current occupation. NOTE: The inter-item reliability (Alpha) for this index is .69. However, because this is a cumulative rather than a correlation index, the items in the index do not necessarily have to be correlated. Rather, as a cumulative index, the items chosen contribute collectively to a conceptual understanding of employment quality.
1. Professional/Managerial Employment in Home Country (Binary Variable)
A score of 1 was given to refugees who were employed in a managerial or professional position, while the remaining respondents received a score of 0.
2. Educational Status Upon Arrival
0. No Schooling
1. Elementary (Grades 1-6)--Incomplete/Complete
2. Junior High (Grades 7-9)--Incomplete/Complete
3. Senior High (Grades 10-l1)--Incomplete
4. Senior High (Grade 12)--Complete
5. Non-university (Vocational/Technical/Nursing) and University--Incomplete
6. Non-university and University (Diploma/Certificate)--Complete
8. Professional Degree or Doctorate--Complete
3. Months of English-language Training
0 = No English-language Training
1 to 54 Months of Training
4. Months of Training in Canada Upon Arrival
0 = No Training in Canada
1 to 48 Months of Training
Social Capital Indices
Proximity to Close Family Ties
1. Do you have adult children living in (current city of residence)?
2. Do you have parents living in (current city of residence)?
3. Do you have siblings living in (current city of residence)?
4. Do you have individuals under 5 years living in your household? (indicating dependents in the home)
5. Do you have individuals between 6 and 14 years living in your household? (indicating dependents in the home)
6. Do you have youths between 15 and 21 years living in your household? (indicating dependents in the home)
Refugees scored a minimum of zero indicating a low proximity to family ties or a maximum of six indicating a high proximity to family ties. One point each was given if refugees had adult children, parents or siblings living in their current city of residence, and dependents between 0-21 years.
Presence of Extra-familial Ties
1. Friends from Home Country
2. Sponsors/Host Volunteer
3. Canadian Friends
5. People from Work (Outside the Workplace)
6. Other Immigrants
Respondents scored one point far each extra-familial contact and could score a maximum of six points (indicating a highly varied extra-familial network) and a minimum of zero (indicating a low presence of extra-familial ties).
Impact of Social Capital in Finding a Job
1. How did you get this (current) job? (open-ended question)
Responses were divided into four categories: "Self" "Social Capital," "Human Capital," "Other Networking." For the purposes of the regression analysis, responses were recoded into a binary variable: "Social Capital" and "Other." The "Other" category includes "Human Capita]," "Other (Forms of Aid)," "Not Employed/Not Applicable/No Response." Respondents were given a score of 1 if they used some form of social capital in finding a job, and a score of zero if they did not.
Table 1a Refugees' Quality of Employment Index % of Respondents Excluding Respondents Out of Total Labour Force and Index Items Respondents Retired Professional/Managerial 7% 9% Part-time Position 28% 35% Temporary Position 30% 37% Received Promotion 25% 31% Overqualified for Current Job 60% 74% Not Satisfied with Current Job 70% 87% Unemployed 16% 20% Total N 525 425 Table 1b Human Capital Human Capital % of Respondents High-school Diploma Received 81% Some Post-secondary Training 50% Professional/Managerial Jobs in Former Home 39% ESL/LINC Training Received in Canada 85% Additional Training Received in Canada 38% Total N 525 Table 2 The Impact of Capital and Control Variables on Quality of Employment R with Quality of Independent Variables Beta Significance Employment Human Capital Occupation in Former Home .08 .155 .003 Education in Former Home -.05 .400 .004 Months of English- -.20 .000 .210 ** language Training Months of Training/ -.10 .041 -.054 Education in Canada Social Capital Proximity to Close -.12 .021 .208 ** Family Members Living With Spouse/Partner .20 .000 .004 Presence of Extra-familial Ties .06 .204 .110 * Familial and Extra-familial .17 .000 .148 ** Aid in Finding a Job Control Variables Sex (Male = 1) .10 .039 .116 * Age -.15 .009 .014 Visible Minority Status (Yes=1) -.14 .012 -.008 Time Spent in Refugee Camp -.01 .849 -.041 Length of Time in Canada -.04 .658 .091 Moved in Canada .07 .223 .038 How Long Lived in Current City .21 .040 .108 * City Size -.05 .286 -.066 Potential Mover -.06 .219 -.085 Difficulty with Credential -.10 .059 -.094 * Recognition Experiences of Discrimination -.11 .039 -.165 * Physical and Psychological -.05 .303 .020 Health Upon Arrival N = 395 [R.sup.2] = .218 * Correlation is significant at the .05 level (2-tailed) ** Correlation is significant at the .01 level (2-tailed)
* I would like to thank Dr. Baha Abu-Laban for allowing me to pursue my dissertation research at the Prairie Contra of Excellence for Research on Immigration and Integration, University of Alberta. This manuscript is based on a segment of my dissertation. My sincere gratitude goes to Dr. Harvey Krahn for his invaluable direction and insights. Finally, I thank Dr. William Carroll and three anonymous reviewers for offering very useful comments and suggestions. This man-
(1.) Borrowing James 5. Coleman's conceptualizations, (1990: 304) human capital entails the skills and knowledge that an individual acquires and uses for future returns. Social capital draws attention to the benefits (not necessarily mutually beneficial) accrued from relationships between individuals. Thus, for a person to possess social capitol he/she must be related to others in some form of social network(s).
(2.) A test for interactions involving gander did not reveal significant effects. That is, the quality of employment showed an improvement for both men and women living with a spouse or partner.
(3.) A test for gender interactions shows that while the negative impact of having young children in the home on quality of employment was stronger for women than for men, this difference is not statistically significant. However, the direction of the results is consistent with previous literature, which shows that preschool children in the home act as barriers to full-time, high-status positions for women, in particular (Eshleman and Wilson, 1998). Perhaps, more refined measures are required, such as directly asking refugees about the barriers to better employment outcomes that are created by the presence of young children in the household (regression results from the author available on request).
Abu-Laban, B., T. Derwing, H. Krahn, M. Mulder and L. Wilkinson. 1999. The Settlement Experiences of Refugees in Alberta. Edmonton: Prairie Centre of Excellence for Research on Immigration and Integration and Population Research Laboratory
Bourdieu, P. 1985. 'The forms of capital." In Handbook of Theory and Research for the Sociology of Education. J.G. Richardson (ed.). New York: Greenwood, pp. 241-58.
Coleman, J.S. 1990. Foundations of Social Theory. Cambridge. Harvard University Press.
Eshleman, J.R. and S.J. Wilson (eds.). 1998. The Family. 2nd ad. Scarborough: Allyn and Bacon.
Giddens, A. 1979. Central Problems in Social Theory. London: Macmillan.
Giddens, A. 1982. Profiles and Critiques in Social Theory. London: Macmillan.
Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration. Cambridge: Polity Press.
Gold, S.J. and N. Kibria. 1993. "Vietnamese refugees and blocked mobility." Asian and Pacific Migration Journal, Vol. 2, No. 1, pp. 27-55.
Haines, D.W (ed.). 1996. Refugees in America in the 1990s. Westport: Greenwood.
Hein, J. 1993. "Refugees, immigrants, and the state." Annual Review of Sociology, Vol. 19, pp. 43-59.
Krahn, H., T. Derwing, M. Mulder and L. Wilkinson. 2000. "Educated and underemployed: Refugee integration into the Canadian labour market." Journal of International Migration and Integration, Vol. 1, No. 1, pp. 59-84.
Krahn, H. and G.S. Lowe. 1998. Work, Industry and Canadian Society. 3rd ad. Toronto: Nelson.
Kunz, J.L., A. Milan and S. Schetagne. 2001. Unequal Access: A Canadian Profile of Racial Differences in Education, Employment and Income. Toronto: The Canadian Race Relations Foundation.
Loury, G.C. 1977. "A dynamic theory of racial income differences." In Women, Minorities and Employment Discrimination. PA. Wallace and A.M. La Mond (ads.). Lexington: Lexington Books, pp. 153-86.
Ohler, S. 1999. "Fast-tracking to a career in nursing: Immigrant doctors get other options." Edmonton Journal, Dec. 2, p. B3.
Pendakur, K. and R. Pandakur. 1997. "Speak and ye shall receive: Language knowledge as human capital." American Journal of Economics and Sociology, Vol. 58, No. 1, pp. 43-56.
Pendakur, K. and R. Pendakur. 1998. "The colour of money: Earnings differentials across ethnic groups in Canada." Canadian Journal of Economics, Vol. 31, No. 3, pp. 518-48.
Piche, V, J. Renaud and L. Gingras. 1997. "Comparative immigrant economic integration." Montreal: Paper Presented at the Second National Metropolis Conference.
Portes, A. 1998. "Social capital: Its origins and applications in modern sociology," Annual Review of Sociology, Vol. 24, pp. 1-24.
Statistics Canada. 2000. "An overview of the 1999 labour market." Labour Force Update, Vol. 4, No. 1, pp. 1-13.…