Public transit use among immigrants.
Subject: Immigrants (Transportation)
Immigrants (Research)
Public transportation (Social aspects)
Public transportation (Statistics)
Authors: Heisz, Andrew
Schellenberg, Grant
Pub Date: 06/22/2004
Publication: Name: Canadian Journal of Urban Research Publisher: Institute of Urban Studies Audience: Academic Format: Magazine/Journal Subject: Social sciences Copyright: COPYRIGHT 2004 Institute of Urban Studies ISSN: 1188-3774
Issue: Date: Summer, 2004 Source Volume: 13 Source Issue: 1
Topic: Event Code: 290 Public affairs; 680 Labor Distribution by Employer; 310 Science & research; 690 Goods & services distribution
Product: Product Code: 9108212 Mass Transit Programs; 9008200 Transportation Programs-Total Govt; E198450 Immigrants NAICS Code: 92612 Regulation and Administration of Transportation Programs
Geographic: Geographic Scope: Canada Geographic Code: 1CANA Canada
Accession Number: 115498151
Full Text: Abstract

This study compares the propensity of immigrants and the Canadian born to use public transit. Using data from the 1996 and 2001 Censuses of Canada, it shows that recent immigrants are much more likely to use public transit to commute to work than Canadian-born persons, even when controlling for demographic characteristics, income, commute distance and residential distance from the city centre. This higher propensity among immigrants falls with time spent in Canada. Recent cohorts of immigrants are more likely to take public transit than earlier cohorts. The implications of these findings for public transit services are discussed in the final section.

Keywords: Immigrants, Public Transit

Resume

A l'aide des donnees des recensements du Canada de 1996 et 2001, nous examinons la tendance des immigrants et des personnes nees au Canada utiliser les transports en commun. Nous remarquons que les recents immigrants sont beaucoup plus susceptibles d'utiliser les transports en commun pour se rendre au travail que les personnes nees au Canada, meme lorsqu'on tient compte des caracteristiques demographiques, du revenu, de la distance a parcourir et de la distance entre le domicile et le centre-ville. Cette forte tendance s'attenue en fonction du temps passe au Canada. Nous remarquons egalement que les recentes cohortes d'immigrants sont plus susceptibles d'utiliser les transports en commun que les cohortes precedentes. On etudie les repercussions sur les services de transport en commun.

Mots ties : Immigrants, Transport en commun

Introduction

Population growth and changes in population composition typically have implications for public services provision. For example, the birth of the baby boom generation was a major factor that contributed to the expansion of the elementary, secondary and post-secondary education systems across Canada in the 1960s. Similarly, there is now ongoing public policy discussion regarding the implications of population aging on health care, pension plans and other programs. Immigration is also an important consideration in this respect. In 2001, the share of Canadians born outside of the country, at 18 percent, was higher than it had been in 70 years. Moreover, immigrants have become increasingly concentrated in Canada's largest cities, with almost three-quarters of "recent immigrants" now residing in Montreal, Toronto and Vancouver. (1) In absolute terms, more than 1.3 million recent immigrants now reside in these three cities, with almost 800,000 of them in Toronto alone. As a result of this trend, the proportion of the population who are recent immigrants rose between 1991 and 2001 from 11.8 to 17 percent in Toronto and from 9.3 to 16.5 percent in Vancouver (Statistics Canada, 2003).

This dramatic shift in population composition towards recent immigrants in Canada's largest cities has important implications for the provision of many public services. This paper examines the importance of increased immigration for the provision of public transit. Specifically, this paper compares the use of public transit to compute to work by immigrants and the Canadian born, and examines whether this difference diminishes with the length of residence in Canada. The central finding is that the propensity to use public transit to commute to work is far higher among recent immigrants than Canadian-born persons and that this difference remains when gender, age, income, distance to work, and residential distance from the city centre are taken into account. One implication is that population growth based on immigration will place greater demands on public transit systems than growth based on natural increase.

The evidence also indicates that immigrants who have been in Canada for more than 20 years typically use public transit to commute to work at the same level as Canadian-born persons. This suggests that either immigrant transit use "integrates" towards the level of the Canadian-born population, and/or that newer cohorts of immigrants have a higher likelihood to use public transit than past cohorts. Our results suggest that both integration and cohort effects are important.

Data

This study uses a descriptive approach to quantify the differences in public transit usage between immigrants and other residents of Canada's cities. (2) The analysis is based on data from the 1996 and 2001 Census micro data files. Specifically, it uses the Census "long form" which contains detailed information from approximately 1 in 5 households. Individuals in these households were asked how they usually get to work, whether they commute by car, truck or van as either a driver or a passenger; by public transit (e.g., bus, streetcar, subway, light rail transit, commuter train, ferry); or by other means, including walking, bicycle, motorcycle or other modes. This paper is only concerned with the use of public transit as the usual mode of transport for getting to work, as the Census does not collect information on public transit use for other types of trips.

Furthermore, our analysis focuses on "recent immigrants," defined as those who arrived in Canada during the 10 years prior to the Census, that is, during the years 1986 to 1996 inclusive for the data derived from the 1996 Census and during the years 1991 to 2001 inclusive for data derived from the 2001 Census. Excluded from our analysis are certain groups, specifically those who were not employed on or after January 1st, 1995 for data derived from the 1996 Census and those not employed on or after January 1st, 2000 for data derived from the 2001 Census; those under 15 years of age; residents of institutions; individuals whose place of work is at home or outside of Canada; and non-permanent residents.

We analyze results at the Census Metropolitan Area (CMA) level and concentrate on three cities: Montreal, Toronto and Vancouver for two reasons: (i) to limit the amount of data presented in the paper and (ii) because almost three-quarters of recent immigrants reside in these three cities. (3) It is important to note that a transit strike was ongoing in Vancouver at the time the 2001 Census was in the field. This strike, which began April 1st, 2001 and ended August 1st, 2001, undoubtedly affected the number of commuters who reported that they usually used public transit to get to work. We discuss the potential effect of the transit strike on our results when it is appropriate.

Immigrants and Public Transit Usage (Descriptive Results)

Although private vehicles remain the primary means by which individuals in Canadian cities commute to work, public transit is an important mode of transportation. Indeed, across Canada's 27 CMAs, more than 1.6 million people usually take public transit to work.

Commuting to work on public transit is especially prevalent among recent immigrants. In Toronto, for example, over one-third (36.3 percent) of recent immigrants, compared with one-fifth (20.7 percent) of Canadian-born persons, usually commute by this means (see Table 1). In Montreal, the share of recent immigrants who usually commute to work on public transit (at 48.6 percent) is more than twice that of Canadian-born persons (at 20.9 percent). A similar pattern is evident in virtually every CMA in which a significant number of recent immigrants reside. In fact, a monotonically declining relationship between years since immigration and public transit usage is observed in all CMAs, with public transit usage rates about the same for immigrants who have resided in Canada for more than 20 years and the Canadian-born.

The difference in rates of commuting by public transit between recent immigrants and the Canadian born can be investigated using a series of thematic maps. (4) Map 1 shows census tracts in Montreal, with each tract shaded according to the share of Canadian-born workers who use public transit to get to work. In most census tracts on the island of Montreal and in Saint Lambert to the east, more than 25 percent of workers in a total of 405 tracts use public transit to get to work. In 34 census tracts, primarily near the downtown core, more than 50 percent of workers take public transit to work. In further outlying areas, typically less than 25 percent of workers take public transit to work.

Map 2 shows the same results for recent immigrants. The difference between the two maps is striking. The number of census tracts (244) in which more than 50 percent of recent immigrants commute via public transit is far higher. Moreover, in areas further from the downtown core, including Beaconsfield and Point Claire to the south and Laval to the north, more than 25 percent of recent immigrants commute via public transit. In short, public transit utilization is higher among recent immigrants than among other groups in Montreal. This is evident in the extent to which utilization rates exceed 50 percent in many of the "central" census tracts and exceed 25 percent in areas further outside the downtown core.

Maps 3 and 4 provide corresponding information for Toronto. As with Montreal, transit usage is highest in the downtown core. In 43 census tracts, more than 50 percent of Canadian-born workers commute to work via public transit; in 305 tracts, between 25 and 50 percent of workers commute in this way. Transit utilization among recent immigrants is again more extensive, as illustrated in Map 4. Indeed more than 50 percent of recent immigrants in a total of 150 census tracts commute via public transit, and between 25 and 50 percent of recent immigrants in another 338 tracts commute in this way. Moreover, public transit utilization is more geographically dispersed. In many census tracts north of Highway 401, in the western cities of Mississauga and Brampton, and the eastern cities of Pickering and Ajax, at least 25 percent of recent immigrants use public transit to commute to work.

[FIGURE 4 OMITTED]

Significantly fewer workers in Vancouver take public transit to work. In only 22 census tracts, most of which are situated along the Sky Train corridor, do more than 25 percent of Canadian-born workers commute in this way (see Map 5). Ridership is more prevalent among recent immigrants, with more than 25 percent of recent immigrants in 74 census tracts commuting in this way. Many of these census tracts are found in North Vancouver, East and Central Vancouver, Surrey and New Westminster, concentrated near routes serviced by the Sky Train and Sea Bus systems. While these numbers were likely affected by the Vancouver transit strike, it remains clear that recent immigrants are more likely to commute on public transit than the Canadian born and that they commute from a more geographically dispersed area.

[FIGURE 5 OMITTED]

Other Correlates of Public Transit Usage

A number of other factors, in addition to immigration status, are associated with commuting on public transit. Descriptive statistics on selected characteristics are provided in Table 2 for persons residing in Toronto, while similar tables for Vancouver and Montreal are provided in the Appendix. Women, whether they are Canadian-born or immigrants, are far more likely than men to commute to work on public transit. However, the gender difference is far larger among recent immigrants. In Toronto, for example, there is a six percentage point difference between the share of Canadian-born women and men who commute to work on public transit, while there is a 17 percentage point difference between immigrant women and men. The same pattern is evident in Vancouver and Montreal (see Appendix Tables 1 and 2). That being said, recent immigrant women are more likely than Canadian-born women to commute using public transit. The same holds true for Canadian-born and immigrant men.

Use of public transit is also correlated with age, as young people are more likely than older individuals to commute to work in this way. Even so, in Toronto, recent immigrants in their forties or fifties are about twice as likely as Canadian-born persons in the same age group to commute via public transit. The same pattern is evident in Vancouver and Montreal.

Economic family income is another factor associated with the likelihood of commuting via public transit. In Toronto, for example, individuals in families with incomes under $40,000 are about twice as likely as those in families with incomes of $80,000 or more to take public transit to work. Through the 1980s and 1990s, immigrants fared quite poorly in the Canadian labour market (Frenette and Morissette, 2003). Consequently, their relatively high rate of public transit utilization may reflect their over-representation at the lower end of the income distribution. However, descriptive statistics do not support this hypothesis. Consider, for example, individuals in Toronto who reside in families with incomes between $60,000 and $80,000. The incidence of public transit utilization among recent immigrants (at 31.7 percent) is 13 percentage points higher than the incidence among Canadian-born persons (at 19.1 percent). Similar differences are evident in other income categories.

The distance that one must travel to work is also associated with the likelihood of commuting via public transit. The likelihood of using public transit increases as the distance to work increases, and then tapers off as the distance to work exceeds about 10 kilometres. However, recent immigrants in Toronto as well as Vancouver and Montreal are far more likely to commute to work using public transit than Canadian-born persons, regardless of the distance to work.

The distance between one's home residence and the city centre also has an impact on the public transit use. Fully 40.1 percent of commuters living within 3 kilometres of Toronto's city centre commute by public transit, while only 12.2 percent of those living 20 kilometres or more from the city centre do. As with other indicators, the likelihood of taking public transit is higher for recent immigrants than other groups for all distances from the downtown core.

Finally, it is important to note that there are significant differences in the utilization rates of public transit among immigrants from different source countries. In both Toronto and Montreal, the incidence of public transit utilization is highest among immigrants from the Caribbean, Southeast Asia, Central and South America, and Africa; they are lowest among immigrants from East Asia, Europe, West Asia, North America and Oceania (See Table 3).

Multivariate Analysis by Cross-section

Having considered the factors associated with public transit utilization in descriptive terms, we now provide a more detailed analysis. This section examines public transit usage among immigrants and others using a logistic model, in which the individual probability of using public transit to commute to work is expressed as a function of individual characteristics and immigration status:

(1) P([Y.sub.i] = [[beta].sub.o] + K[summation over (k=1)] [[beta].sub.1k] [X.sub.ki] + C[summation over (c=1)] [[beta].sub.2c] [I.sub.ci] + [[epsilon].sub.i]

In equation (1) Y denotes whether the individual (i) takes public transit. Included in X, the first component of the model, is a series of background characteristics which are assumed to influence the probability of taking public transit. These include gender, age, economic family income, distance traveled to work, and distance from place of residence to the city centre. The second component of the model, I, represents a series of three immigrant dummy variables indicating whether the immigrant is a recent immigrant (0-10 years before the Census year), a medium-term immigrant (11 to 20 years in Canada), or a long-term immigrant (more than 20 years in Canada). The reference group is the Canadian born. The dummy variables for the three terms (of years since immigration) are also interacted with age, economic family income, distance to work, and distance from the city centre to allow for the impacts of these factors to vary for different groups of immigrants. The error term [epsilon] is assumed to be randomly distributed. Models are estimated separately for Montreal, Toronto and Vancouver.

Table 4 shows the results of the regressions for public transit usage. Column 1 shows results for Montreal, where immigrants remain more likely to commute by public transit after controlling for gender, age, family income and distance to work. Estimates indicate that, compared to Canadian-born persons in Montreal, recent immigrants are 1.91 times more likely to take public transit to work; medium-term immigrants are 1.46 times more likely; and long-term immigrants are 1.06 times more likely. The model confirms that transit usage declines with age, higher family income and greater distance traveled to work and is lower for men than women. Interaction terms between these variables and immigration status indicate that, while transit use declines with age and distance to work for all groups, they decline less for recent immigrants than Canadian-born persons. This means that, relative to their Canadian-born counterparts, immigrants in Montreal are relatively more likely to use public transit as they get older or reside further from work. In contrast, interaction terms with income suggest that public transit usage declines faster among recent immigrants than among Canadian-born persons as family income rises. This may reflect a number of factors, such as a greater preference to substitute private transportation for public transit as income increases or different choices in where immigrants and Canadian-born persons in various income categories decide to live. The gender-immigration interaction terms are strong and significant, confirming our earlier finding that gender differences related to the likelihood of taking public transit are larger among immigrants than the Canadian born in Montreal. Finally, the distance between residence and city centre interaction terms indicate that, for Montreal, living further from the downtown core is less of a factor for public transit use among recent immigrants than the Canadian born. (5)

The results of the logit regressions for Toronto and Vancouver (shown in Columns 2 and 3 in Table 4) are very similar in qualitative terms to those for Montreal. Estimates suggest that recent immigrants in Toronto are 1.46 times more likely to commute to work using public transit than Canadian-born persons, while medium-term immigrants are 1.20 times more likely and long-term immigrants 1.03 times more likely. In Vancouver, recent immigrants are 1.64 times more likely to use public transit to commute and medium-term immigrants are 1.14 times more likely. The difference between long-term immigrants and the Canadian born is not significant. It was earlier shown that Montreal commuters are less likely to use public transit when they must travel longer distances to work. The opposite is true in Toronto and Vancouver, as the likelihood of commuting via public transit is positively correlated with distance to work.

To help interpret the results in Table 4, predicted probabilities of commuting on public transit for recent immigrants and Canadian-born persons with various age, income and distance to work values are presented in Table 5. For example, consider a male commuter in Montreal, aged 25, with an economic family income of $50,000, who travels 5 kilometres to work and lives 10 kilometres from the city centre. A Canadian-born male with these characteristics has a predicted probability of 0.29 of taking public transit, while a recent immigrant with the same characteristics has a predicted probability of 0.43 or 47 percent greater probability to take public transit than the Canadian born. The probability of commuting via public transit is lower for workers aged 35 (who otherwise have the same characteristics) but falls less for recent immigrants. Consequently, a 35-year-old immigrant with these same characteristics is about 60 percent more likely than his Canadian-born counterpart to take public transit. Increasing income to $75,000, and distance traveled to work to 10 kilometres, has relatively little impact on the transit use of commuters. Doubling the distance from home residence to the city centre from 10 to 20 kilometres decreases commute rates for both Canadian-born persons and recent immigrants but the magnitude of the decline is smaller among recent immigrants.

The major details of this story are essentially the same for Toronto and Vancouver. Recent immigrants appear much more willing to use public transit to commute to work than the Canadian born. This remains true whether they are older, reside in families with higher incomes, travel further to work or live further from the city centre. (6)

Multivariate Analysis by Cohort

The results presented in Table 4 imply that the likelihood of commuting on public transit declines as immigrants reside in Canada for longer periods of time. This may reflect rising incomes as individuals settle into the labour market, the acquisition of English or French, the attainment of a Canadian driver's license, or a shift in values or preferences. Such factors may be termed "integration effects." Alternatively, immigrants arriving in Canada in more recent years may be more likely to take public transit than those who arrived before them for other reasons, resulting from changes in source regions or other characteristics common to members of the new cohort. These may be termed "cohort effects." We assess the relative importance of these effects with the model:

(2) P([Y.sub.i] = 1) = [[beta].sub.o] + K[summation over (k=1] [[beta].sub.1k] [X.sub.ki] + R[summation over (r=1)] [[beta].sub.2r] [S.sub.ri] + [[beta].sub.3][T.sub.i] + [[beta].sub.4][C.sub.i] + [[epsilon].sub.i]

This model only includes immigrants and pools data from both the 1996 and 2001 Censuses. The pooling of census files from different years allows us to separate the effects of when an individual immigrated (the cohort effect) from how long the individual has resided in Canada (the integration effect), since cohorts are observed at two points in time. As in Model 1, the first component of the model, X, includes gender, age, economic family income, distance to work and distance from place of residence to the city centre. The second component, S, represents a series of source region dummy variables. These dummy variables control for the possibility that immigrants from different regions have different propensities to commute on public transit. This allows us to take the changing source region composition of immigrant cohorts into account. The third component, T, is a linear term that denotes the number of years the immigrant has resided in Canada. The fourth component, C, is also a linear term that denotes the immigration cohort (where C=immigration year). A negative coefficient for [[beta.sub.3] indicates that the probability of taking public transit to work declines as immigrants live in Canada longer (the integration effect). A positive coefficient for [[beta.sub.4] indicates that more recent cohorts of immigrants have higher probabilities of commuting on public transit (the cohort effect). Because we are using only two census cross-sections for this analysis, it is not possible to separate the effects of years since immigration from the possible effects of the Vancouver transit strike. (7) Hence, we do not report results for Vancouver.

Results are presented in Table 6 for Montreal and Toronto. Column 1 shows results from the first two components of the model (that is, gender, age, income, distance to work, and distance from the city centre as well as years since arrival). It confirms that the likelihood of commuting to work on public transit declines as immigrants reside in Canada for longer periods of time. Column 2 shows the results when source region variables are added. To simplify the table, we have not presented the results for the source region dummy variables. However, it should be noted that many of these variables are statistically significant, indicating important differences for immigrants from different regions. Most importantly, the coefficients associated with years since arrival change very little from Columns 1 and 2. This suggests that it is not a simple shift in source region that underlies the higher transit use among more recent immigrants. Column 3 adds the immigration cohort variable. This variable enters the model significantly in both CMAs, reducing the years since immigration effect in Montreal by about 50 percent and in Toronto by about 25 percent. This indicates that about one-half of the higher public transit usage rate among recent immigrants in Montreal is associated with rates of higher usage among newer arrival cohorts, independent of how long they have resided in Canada. In Toronto, about one-quarter of the higher rate of public transit usage among more recent cohorts is associated with a cohort effect, while the remainder is associated with an integration effect.

To put these results in more concrete terms, we computed predicted values from the regressions in Column 3 of Table 6. (8) Predicted values from the model suggest that, one year after immigration, 34.5 percent of Montreal workers who immigrated in 1990 commuted on public transit. Ten years later, only 30.7 percent from this cohort still commute to work. This indicates a considerable integration effect. However, one year after immigration, 37.1 percent of Montreal workers who immigrated in 2000 commuted to work, indicating that the 2000 cohort of arrivals are 2.6 percentage points more likely to take public transit than the 1990 cohort (measured one year after immigration). Toronto immigrants from the 2000 cohort are 1.1 percentage points more likely to take public transit than immigrants in the 1990 cohort. (9)

Conclusion

In this study, we examine the rate of public transit use among immigrant workers in Montreal, Toronto and Vancouver, using Census data from 1996 and 2001. Comparisons are drawn with Canadian-born workers as well as across immigrant cohorts. We find that recent immigrants are much more likely than the Canadian born to use public transit to commute to work, even after controlling for age, gender, income, distance to work, and distance between place of residence and the city centre. Two factors seem to explain this high rate of transit usage. First, immigrants tend to use public transit in their commute to work more when they are new to Canada (independent of other factors such as age and income) but their rate of transit use declines as they reside in Canada for longer periods of time. Second, newer cohorts of immigrants have higher rates of transit use than earlier cohorts, suggesting that they may be different in some ways that have not been observed.

The high rate of public transit utilization among recent immigrants, coupled with the fact that immigration has become the most important source of population growth in many CMAs, has two important implications for public transit. First, projections for future public transit needs could take into account that the urban population is not only growing, but is also compositionally shifting towards a high-usage group. Second, immigrants have a high-usage rate no matter how far away they live from the downtown core. Unlike earlier cohorts who initially settled in the downtown areas of CMAs, many immigrants in the 1980s and 1990s have settled directly in suburban areas (Balakrishnan and Hou, 1999; Myles and Hou, 2003). A shift in the geographic concentration of immigrants from urban core to outlying areas has implications for the location of public transit services, especially in CMAs with centralized transit systems.

Acknowledgements

The authors would like to thank Feng Hou for assistance in constructing the data and for comments on an earlier draft.

Notes

(1) Recent immigrants are defined as those who arrived in Canada during the 10 years prior to the Census.

(2) While the present paper is concerned only with descriptive differences, it should be noted that the issue of inter-group differences in public transit usage typically arises in one of two contexts: (1) the modal choice literature, which is concerned with quantifying the determinants mode of commuting choice (see Ascension, 2002, for a recent example); and (2) the literature regarding spatial mismatch in access to employment which examines the commuting times of women, immigrants and ethnic minorities, relative to other groups, and the role this plays in labour market decisions (see, for example, McLafferty and Preston, 1996; Zax, 1998; and Preston, McLafferty and Liu, 1998).

(3) A small number of observations where distance to work was reported as zero kilometers was also dropped. Definitions of Census geographic and other concepts used in this paper are available at www.statcan.ca.

(4) For this analysis, the dataset is further restricted to those Census tracts with a sample population of more than 250 persons. Public transit usage rates among recent immigrant commuters are only reported when there is a sample population of more then 40.

(5) Other models were estimated which allowed the impact of age, economic family income and distance to work to enter the model non-linearly. While the higher order terms were often significant, they yielded little additional insight to the results and so are not reported here.

(6) Results that discuss relative differences across groups at a point in time should not be affected by the Vancouver transit strike to the extent that the strike affected all groups in a proportionate manner. Furthermore, analyses using 1996 and 2001 Census data yield similar findings, indicating that the 2001 cross-sectional results are robust to the influence of the Vancouver transit strike.

(7) Information on mode of transportation to work was not collected prior to the 1996 Census.

(8) These predictions are for men, with economic family income of $50,000, commuting 5 kilometres to work and living 10 kilometres from the city centre. They were aged 24 at the time of immigration. For comparison purposes, a Canadian-born person with the same characteristics residing in Montreal is predicted to have a 24 percent probability of commuting on public transit at age 35 in 2001.

(9) Predicted results from the Toronto model suggest that, one year after immigration, 37.1 percent of workers from the 1990 cohort took public transit to work, compared to 38.2 percent of workers from the 2000 cohort.

References

Balakrishnan, T.R. and Feng Hou. 1999. Socioeconomic integration and spatial residential patterns of immigrant groups in Canada. Population Research and Policy Review 18:201-217.

Frenette, Marc and Rene Morissette. 2003. Will they ever converge? Earnings of immigrants and Canadian-born workers over the last two decades. Analytical Studies Branch Research Paper Series. Paper No. 215. Statistics Canada, Ottawa.

Hanson, S. and G. Pratt. 1995. Gender, Work and Space. New York: Routledge.

Kalbach, W. E. 1990. Ethnic residential segregation and its significance for the individual in an urban setting. In Ethnic Identity and Equality, ed. R. Breton et al. Toronto: University of Toronto Press.

McLafferty, S. and V. Preston. 1996. Gender, race and the determinants of commuting: New York in 1990. Urban Geography 16: 192-212.

Myles, John and Feng Hou. 2003. Neighbourhood attainment and residential segregation among Toronto's visible minorities. Analytical Studies Branch Research Paper Series. Paper No, 206. Statistics Canada, Ottawa.

Preston, Valerie, S. McLafferty, and X. F. Liu. 1998. Geographical barriers to employment for American-born and immigrant workers. Urban Studies. 35 (3): 529-545.

Statistics Canada. 2003. Update on cultural diversity. Canadian Social Trends. (Autumn).

Zax, Jeffrey S. 1998. Immigration, Race and Space. In Help or Hindrance? The Economic Implications of Immigration for African Americans, ed. D. Hamermest et al. New York: Russell Sage Foundation.
Appendix

Table 1. Percent of persons in Vancouver who use public transportation
to commute to work, by immigration status and selected characteristics,
2001

                                          Immigrated       Immigrated
                           Canadian         0 to 10         11 to 20
                             born        years earlier    years earlier

Total                       11.4%            21.1%            14.7%

Gender
  Men                        9.6%            15.7%            10.0%
  Women                     13.2%            26.2%            19.1%

Age Group
  15 to 29                  15.0%            23.8%            18.4%
  30 to 39                  10.6%            21.1%            12.7%
  40 to 49                   9.2%            17.6%            12.8%
  50 to 59                   8.6%            19.6%            15.2%

Family Income
  $1 to $19,999             22.5%            25.4%            20.8%
  $20,000 to $39,999        16.9%            21.7%            17.1%
  $40,000 to $59,999        12.3%            21.8%            15.4%
  $60,000 to $79,999        10.0%            20.1%            13.8%
  $80,000 or more            7.9%            17.5%            12.4%

Distance to Work
  Less than 5 kms            9.8%            18.5%            12.5%
  5 to 9.9 kms              13.2%            24.2%            17.4%
  10 to 19.9 kms            10.7%            21.7%            14.3%
  20 kms or more            13.3%            21.2%            15.4%

Distance from City
Centre

  0 to 2.9 km               36.6%            46.3%            42.7%
  3 to 5.9 km               40.2%            58.5%            47.4%
  6 to 9.9 km               31.6%            51.6%            40.5%
  10 to 14.9 km             22.3%            46.8%            34.5%
  15 to 19.9 km             16.1%            27.0%            22.1%
  20 or more                 7.3%            22.8%            15.9%

                          Immigrated
                         more than 20
                         years earlier       Total

Total                       11.0%            13.1%

Gender
  Men                        7.6%            10.2%
  Women                     14.6%            15.9%

Age Group
  15 to 29                  16.1%            16.7%
  30 to 39                  11.5%            13.1%
  40 to 49                   9.9%            11.0%
  50 to 59                  10.7%            10.6%

Family Income
  $1 to $19,999             21.8%            23.1%
  $20,000 to $39,999        15.6%            17.9%
  $40,000 to $59,999       131.0%            14.4%
  $60,000 to $79,999        10.8%            11.8%
  $80,000 or more            8.2%             9.1%

Distance to Work
  Less than 5 kms            8.7%            11.3%
  5 to 9.9 kms              12.9%            15.4%
  10 to 19.9 kms            10.3%            12.6%
  20 kms or more            13.9%            14.4%

Distance from City
Centre

  0 to 2.9 km               32.1%            37.6%
  3 to 5.9 km               36.8%            42.5%
  6 to 9.9 km               29.6%            34.1%
  10 to 14.9 km             24.8%            24.5%
  15 to 19.9 km             15.8%            16.7%
  20 or more                10.2%             7.8%

Table 2. Percent of persons in Montreal who use public transportation
to commute to work, by immigration status and selected characteristics,
2001

                                          Immigrated       Immigrated
                           Canadian         0 to 10         11 to 20
                             born        years earlier    years earlier

Total                       20.9%            48.6%            35.5%

Gender
  Men                       16.9%            41.0%            26.9%
  Women                     24.7%            57.7%            45.7%

Age Group
  15 to 29                  30.5%            58.6%            49.0%
  30 to 39                  16.9%            45-0%            28.9%
  40 to 49                  16.8%            42.2%            31.8%
  50 to 59                  15.9%            42.7%            31.5%

Family Income
  $1 to $19,999             35.8%            61.5%            53.6%
  $20,000 to $39,999        27.8%            51.2%            41.4%
  $40,000 to $59,999        20.1%            44.7%            34.2%
  $60,000 to $79,999        17.1%            39.4%            29.3%
  $80,000 or more           15.8%            35-8%            25.3%

Distance to Work
  Less than 5 kms           20.2%            46.6%            35.4%
  5 to 9.9 kms              30.4%            58.1%            41.8%
  10 to 19.9 kms            18.6%            44.5%            32.0%
  20 kms or more            11.9%            26.2%            20.3%

Distance from
City Centre
  0 to 2.9 km               15.5%            21.3%            15.8%
  3 to 5.9 km               18.8%            29.0%            22.0%
  6 to 9.9 km               15.7%            25.9%            19.0%
  10 to 14.9 km             12.5%            20.9%            13.3%
  15 to 19.9 km             13.4%            21.7%            15.8%
  20 or more                 6.9%            13.6%             9.0%

                          Immigrated
                         more than 20
                         years earlier       Total

Total                       24.5%            23.6%

Gender
  Men                       16.2%            19.0%
  Women                     34.6%            28.3%

Age Group
  15 to 29                  37.3%            33.2%
  30 to 39                  23.0%            20.7%
  40 to 49                  23.0%            19.5%
  50 to 59                  24.5%            18.7%

Family Income
  $1 to $19,999             40.2%            40.5%
  $20,000 to $39,999        34.7%            31.5%
  $40,000 to $59,999        24.9%            22.8%
  $60,000 to $79,999        23.3%            19.0%
  $80,000 or more           17.3%            16.8%

Distance to Work
  Less than 5 kms           22.9%            23.2%
  5 to 9.9 kms              31.0%            33.3%
  10 to 19.9 kms            22.9%            20.8%
  20 kms or more            14.4%            12.6%

Distance from
City Centre
  0 to 2.9 km               15.0%            16.2%
  3 to 5.9 km               15.5%            20.2%
  6 to 9.9 km               13.5%            17.7%
  10 to 14.9 km             10.5%            14.2%
  15 to 19.9 km             11.0%            14.6%
  20 or more                 7.6%             7.8%

Table 1. Percent of persons who use public transportation to commute to
work, by immigration status, selected CMAs, 2001

                                 Immigrated      Immigrated
                                 Within past      11 to 20
               Canadian-born      10 years      years earlier

Montreal           20.9%            48.6%           35.5%
Toronto            20.7%            36.3%           26.8%
Ottawa-Hull        18.5%            33.8%           26.5%
Calgary            13.6%            25.8%           17.0%
Winnipeg           14.1%            24.5%           16.4%
Vancouver          11.4%            21.1%           14.7%
Edmonton            9.4%            19.6%           11.8%
Victoria           10.8%            18.3%           13.9%
Hamilton            8.5%            16.7%           10.8%
London              6.8%            15.1%            8.0%
Windsor             3.3%             9.8%            5.3%
Kitchener           4.5%             9.0%            6.1%

                 Immigrated
                 more than
                20 years ago        Total

Montreal           24.5%            23.6%
Toronto            19.9%            24.0%
Ottawa-Hull        18.9%            19.9%
Calgary            13.5%            14.8%
Winnipeg           12.7%            14.5%
Vancouver          11.0%            13.1%
Edmonton            7.8%             9.9%
Victoria            8.9%            11.0%
Hamilton            6.5%             8.8%
London              4.0%             6.9%
Windsor             1.8%             3.6%
Kitchener           2.5%             4.6%

Table 2. Percent of persons in Toronto who use public transportation to
commute to work, by immigration status and selected characteristics,
Toronto 2001

                          Canadian       Immigrated      Immigrated
                            born          0 to 10         11 to 20
                                        years earlier   years earlier

Total                      20.7%           36.3%            26.8%

Gender
  Men                      17.7%           28.2%            18.3%
  Women                    23.5%           44.8%            35.3%

Age Group
  15 to 29                 26.5%           42.8%            35.6%
  30 to 39                 19.6%           34.8%            23.3%
  40 to 49                 16.3%           31.5%            23.6%
  50 to 59                 14.6%           32.8%            25.5%

Family Income
  $1 to $19,999            37.9%           49.5%            41.5%
  $20,000 to $39,999       32.7%           41.8%            35.7%
  $40,000 to $59,999       24.2%           35.2%            28.1%
  $60,000 to $79,999       19.1%           31.7%            25.3%
  $80,000 or more          16.4%           28.9%            21.8%

Distance to Work
  Less than 5 kms          19.8%           34.8%            25.2%
  5 to 9.9 kms             27.0%           38.9%            28.1%
  10 to 19.9 kms           19.5%           38.7%            28.2%
  20 kms or more           16.5%           29.7%            25.1%

Distance from City
Centre
  0 to 2.9 km              36.9%           52.4%            43.0%
  3 to 5.9 km              45.0%           58.8%            49.1%
  6 to 9.9 km              35.5%           51.7%            43.0%
  10 to 14.9 km            27.8%           43.9%            35.9%
  15 to 19.9 km            21.9%           35.1%            27.5%
  20 or more               10.0%           20.9%            15.1%

                         Immigrated        Total
                        more than 20
                        years earlier

Total                      19.9%           24.0%

Gender
  Men                      12.9%           18.8%
  Women                    27.0%           29.3%

Age Group
  15 to 29                 32.3%           30.6%
  30 to 39                 22.4%           24.2%
  40 to 49                 18.3%           20.5%
  50 to 59                 19.1%           19.2%

Family Income
  $1 to $19,999            34.0%           42.1%
  $20,000 to $39,999       31.0%           35.6%
  $40,000 to $59,999       23.4%           27.2%
  $60,000 to $79,999       19.8%           22.3%
  $80,000 or more          16.0%           18.1%

Distance to Work
  Less than 5 kms          17.4%           22.8%
  5 to 9.9 kms             21.3%           28.5%
  10 to 19.9 kms           20.8%           24.6%
  20 kms or more           19.8%           19.7%

Distance from City
Centre
  0 to 2.9 km              34.2%           40.1%
  3 to 5.9 km              40.1%           46.8%
  6 to 9.9 km              33.2%           39.0%
  10 to 14.9 km            25.7%           32.3%
  15 to 19.9 km            20.8%           26.1%
  20 or more               11.0%           12.2%

Table 3. Percent of recent immigrants who use public transportation to
commute to work, by region of birth, selected CMAs, 2001

                                       Toronto    Vancouver    Montreal

  Caribbean                               47.2%       22.4%       60.4%
  Southeast Asia                          44.9%       30.1%       51.9%
  Central & South America                 42.5%       29.1%       53.5%
  Africa                                  41.2%       21.1%       58.1%
  South Asia                              35.2%      168.0%       51.2%
  Eastern Europe                         329.0%      239.0%       41.4%
  Northern, Western & Southern Europe     31.5%       18.0%       38.4%
  East Asia                               30.8%       19.1%       42.1%
  West Asia                               29.1%       23.4%       36.9%
  North America, Oceania & Other         272.0%      133.0%       29.7%
Total Recent Immigrants                   36.3%       21.1%       48.6%

Canadian-born                             20.7%       11.4%       20.9%

Table 4: Logistic regression results, probability of commuting via
public transit

                                                         Montreal
                                                            (1)

Constant                                                  -0.228 *
                                                          (0.011)

immigrated 0-10 years ago                                  0.439 *
                                                          (0.040)

immigrated 11-20 years ago                                 0.304 *
                                                          (0.046)

immigrated more than 20 years ago                         -0.047
                                                          (0.042)

Age                                                       -0.274 *
                                                          (0.005)

Female                                                     0.497 *
                                                          (0.012)

economic family income                                    -0.040 *
                                                          (0.001)

distance to work                                          -0.011 *
                                                          (0.002)

distance from city centre                                 -0.088 *
                                                          (0.001)

immigrated 0-10 years ago * economic family income        -0.043 *
                                                          (0.006)
immigrated 11-20 years ago * economic family
income                                                    -0.030 *
                                                          (0.006)

immigrated more than 20 years ago * economic              -0.029 *
family income                                             (0.004)

Immigrated 0-10 years ago * age                            0.070 *
                                                          (0.019)

immigrated 11-20 years ago * age                           0.064 *
                                                          (0.021)

immigrated more than 20 years ago * age                    0.279 *
                                                          (0.018)

immigrated 0-10 years ago * distance to work              -0.001
                                                          (0.005)

immigrated 11-20 years ago * distance to work              0.010
                                                          (0.007)

immigrated more than 20 years ago * distance to work       0.013 *
                                                          (0.005)

immigrated 0-10 years ago * female                         0.225 *
                                                          (0.036)

immigrated 11-20 years ago * female                        0.399 *
                                                          (0.044)

immigrated more than 20 years ago * female                 0.520 *
                                                          (0.037)

immigrated 0-10 years ago * distance from city centre      0.027 *
                                                          (0.003)
immigrated 11-20 years ago * distance from city
centre                                                     0.012 *
                                                          (0.004)

immigrated more than 20 years ago * distance from          0.010 *
city centre                                               (0.003)

N                                                          308276
log likelihood                                            -144547

                                                          Toronto
                                                            (2)

Constant                                                  -0.194 *
                                                          (0.011)

immigrated 0-10 years ago                                  0.551 *
                                                          (0.025)

immigrated 11-20 years ago                                 0.327 *
                                                          (0.030)

immigrated more than 20 years ago                          0.034
                                                          (0.026)

Age                                                       -0.235 *
                                                          (0.005)

Female                                                     0.367 *
                                                          (0.012)

economic family income                                    -0.023 *
                                                          (0.001)

distance to work                                           0.007 *
                                                          (0.002)

distance from city centre                                 -0.067 *
                                                          (0.001)

immigrated 0-10 years ago * economic family income        -0.020 *
                                                          (0.003)
immigrated 11-20 years ago * economic family
income                                                    -0.005
                                                          (0.003)

immigrated more than 20 years ago * economic              -0.012 *
family income                                             (0.003)

Immigrated 0-10 years ago * age                            0.091 *
                                                          (0.010)

immigrated 11-20 years ago * age                           0.090 *
                                                          (0.013)

immigrated more than 20 years ago * age                    0.116 *
                                                          (0.011)

immigrated 0-10 years ago * distance to work               0.002
                                                          (0.004)

immigrated 11-20 years ago * distance to work              0.021 *
                                                          (0.005)

immigrated more than 20 years ago * distance to work       0.036 *
                                                          (0.005)

immigrated 0-10 years ago * female                         0.418 *
                                                          (0.022)

immigrated 11-20 years ago * female                        0.586 *
                                                          (0.027)

immigrated more than 20 years ago * female                 0.576 *
                                                          (0.025)

immigrated 0-10 years ago * distance from city centre     -0.003+
                                                          -0.001
immigrated 11-20 years ago * distance from city
centre                                                    -0.005*
                                                          (0.002)

immigrated more than 20 years ago * distance from          0.000
city centre                                               (0.001)

N                                                          425645
log likelihood                                            -204438

                                                         Vancouver
                                                            (3)

Constant                                                  -1.502 *
                                                          (0.018)

immigrated 0-10 years ago                                  0.513 *
                                                          (0.039)

immigrated 11-20 years ago                                 0.197 *
                                                          (0.050)

immigrated more than 20 years ago                         -0.174 *
                                                          (0.047)

Age                                                       -0.157 *
                                                          (0.010)

Female                                                     0.343 *
                                                          (0.023)

economic family income                                    -0.057 *
                                                          (0.003)

distance to work                                           0.012 *
                                                          (0.002)

distance from city centre                                 -0.039 *
                                                          (0.001)

immigrated 0-10 years ago * economic family income         0.025 *
                                                          (0.005)
immigrated 11-20 years ago * economic family
income                                                     0.014+
                                                          (0.007)

immigrated more than 20 years ago * economic              -0.009
family income                                             (0.007)

Immigrated 0-10 years ago * age                            0.088 *
                                                          (0.020)

immigrated 11-20 years ago * age                           0.104 *
                                                          (0.027)

immigrated more than 20 years ago * age                    0.151 *
                                                          (0.023)

immigrated 0-10 years ago * distance to work               0.002
                                                          (0.004)

immigrated 11-20 years ago * distance to work             -0.007
                                                          (0.005)

immigrated more than 20 years ago * distance to work      -0.002
                                                          (0.004)

immigrated 0-10 years ago * female                         0.301 *
                                                          (0.043)

immigrated 11-20 years ago * female                        0.412 *
                                                          (0.060)

immigrated more than 20 years ago * female                 0.355 *
                                                          (0.052)

immigrated 0-10 years ago * distance from city centre      0.004
                                                          (0.002)
immigrated 11-20 years ago * distance from city
centre                                                    -0.002
                                                          (0.003)

immigrated more than 20 years ago * distance from          0.008 *
city centre                                               (0.002)

N                                                         165898
log likelihood                                            -60424

Note: An asterisk beside a coefficient indicates that it is
Statistically significant at the 5% level.

Table 5: predicted probabilities of commuting via public transit, Men

           economic                          distance
            family          distance        from city
Age         income          to work           centre

                                             Montreal
25           50000             5                10
35           50000             5                10
25           75000             5                10
25           50000             10               10
25           50000             5                20

                                             Toronto
25           50000             5                10
35           50000             5                10
25           75000             5                10
25           50000             10               10
25           50000             5                20

                                            Vancouver
25           50000             5                10
35           50000             5                10
25           75000             5                10
25           50000             10               10
25           50000             5                20

           predicted probability of commuting via
                            public transit *

           Canadian-        recent-
Age          born           immigrant         ratio

25           0.29             0.43             1.47
35           0.24             0.38             1.59
25           0.27             0.38             1.39
25           0.29             0.42             1.47
25           0.14             0.29             1.98

25           0.35             0.41             1.19
35           0.29             0.38             1.28
25           0.33             0.38             1.16
25           0.35             0.41             1.19
25           0.21             0.26             1.21

25           0.16             0.18             1.18
35           0.14             0.17             1.28
25           0.14             0.17             1.25
25           0.16             0.19             1.18
25           0.11             0.14             1.23

* based upon regression results reported in table 4

Table 6: Cohort and Integration Effects, Montreal and Toronto

                                            Montreal
                                  (1)          (2)         (3)

Time since immigration          -0.023 *    -0.021 *     -0.010 *
                                (0.001)      (0-001)     (0.003)

Cohort of immigration                                    0.011 *
                                                         (0.003)

N                                111443      111443       111443
log-likelihood                   -63052      -62200       -62191

                                             Toronto

Time since immigration          -0.024 *    -0.021 *     -0.016 *
                                (0.000)      -0.001      (0.002)

Cohort of immigration                                    0.005 *
                                                         (0.002)

N                                383904       383904      383904
log-likelihood                  -197893      -195099     -195095

Constant, age, female, distance to work, and distance
from city centre controls         Yes          Yes         Yes

Country of origin controls         No          Yes         Yes

Note: An asterisk beside a coefficient indicates that it is
statistically significant at the 1% level
Gale Copyright: Copyright 2004 Gale, Cengage Learning. All rights reserved.