Trends in Social Capital in Canada

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

By Martin Turcotte

[Highlights][Full article in PDF]

Skip to text

• Section 1: Trends in social networks
• Section 2: Trend in civic engagement and volunteerism
• Section 3: Trends in feelings of trust
• Conclusion
• Data sources
• References
• Notes

Text begins

There is rarely consensus on social science concepts and the ‘social capital’ concept is no exception. In general, social capital is considered a set of resources available to individuals and communities as a result of social networks. The value of social capital resides in the overall positive consequences resulting from the creation and maintenance of social contacts: flow of information, trust, reciprocity, co-operation, and productivity.^Note 1

There are two broad approaches within the study of social capital. The first approach regards social capital as an individual resource mainly benefiting the members of social networks (Bourdieu 1986; Lin 2001; Coleman 1988; Boxman et al. 1991).^Note 2 According to this individualist approach, a high level of social capital is defined by (1) strong, reliable personal networks based on reciprocity, that allow individuals to get emotional support, companionship and financial assistance in an emergency, and which are referred to as ‘strong ties’; and (2) extended and diverse networks of acquaintances that provide access to key resources (privileged information, contacts, job opportunities, etc.),^Note 3 and which are referred to as ‘weak ties.’

According to certain proponents of the individualist approach, individuals ‘invest’ in their social relationship networks and draw future benefits from them, even if it is not their main goal. This process is similar to individuals investing in education to increase their human capital, their employability and their long‑term income (Lin 2001). The individual benefits of social capital, which are often considered public policy goals, are many, including positive health effects (Berkman and Glass 2000), integration of immigrants into the labour market (Chiswick and Miller 1996), and quality of employment and income levels (Erickson 2001). By definition, this individualist view of social capital underlies the existence of inequalities, since many people are not members of social networks that provide access to the scarcest and most desired resources (Bourdieu 1986).

The second approach argues that social capital is a characteristic of communities, whether it is neighbourhoods, cities, regions or countries (Putnam 2000). Despite this more collective approach to understanding social capital, this perspective still considers social networks as the source of social capital, particularly through involvement and participation in the community. Community involvement and social contacts give rise to the standards, values and behaviours that benefit the whole of society. According to this view, one of the fundamental characteristics of communities with a high level of social capital is the tendency of citizens to trust one another—even if they do not know each other—which is referred to as ‘generalized trust.’ In short, even though social capital or ‘generalized trust’ can be the product of the diverse relationships and social contacts between individuals, the emphasis is on the positive effects for society as a whole.

Previous studies, based on the collective approach, have found that when a community has high levels of social capital, crime levels are lower, children perform better in school, political involvement is higher, and economic growth is stronger (for examples, see Kay and Johnson 2007).

The purpose of this report is not to discuss the various theoretical debates between social capital experts, but rather to illustrate trends in the various indicators and measures identified in previous studies.

Statistics Canada first included a number of measures associated with the concept of social capital in its 2003 General Social Survey (GSS) on Social Engagement. The 2008 and 2013 GSS also included these measures. It is now possible to present a detailed portrait of trends in these indicators over a decade.

Section 1 of the report focuses on indicators that mainly stem from the approach that views social capital as an individual characteristic: size of networks of close friends; networks of acquaintances, neighbours and other friends; frequency of contacts; and diversity of friends. Sections 2 and 3 present indicators mostly associated with the collective approach. Section 2 examines how Canadians’ participation in organizations or associations evolved between 2003 and 2013. Section 3 deals with a key measure of the social capital concept, namely generalized trust. It also presents various other indicators of people’s trust in the people in their neighbourhood and in strangers.

Section 1: Trends in social networks^Note 4

Seniors especially at risk of having no close friends

Close friends are “people who are not your relatives, but who you feel at ease with, can talk to about what is on your mind, or call on for help”. These individuals are the ones who provide emotional or financial help during times of difficulty and who often know the person best (Wellman and Wortley 1990).

It is difficult, if not impossible, to determine the ‘ideal’ or desirable number of close friends, particularly because there is wide variation in individual preferences. That being said, having no close friends is probably not a desirable situation for most people.^Note 5 In 2013, a small minority of Canadians reported having no close friends (6%)^Note 6. The trend was relatively stable, as this proportion is virtually unchanged from 2003 (Table 1).

Despite this small proportion of Canadians with no close friends, the proportion varies significantly by age group. In 2013, for example, 15% of seniors aged 75 years and older were in this situation, compared with 5% of persons aged 35 to 44. In other words, among Canadians aged 75 and older not living in an institution, almost 1 in 6 had no close friends they could confide in or call on for help.

According to certain studies, if these seniors’ lack of friendships translates into a feeling of loneliness, it could be a risk factor in the loss of functional capacity and even death (Perissinotto et al. 2012; Luo et al. 2012). According to the 2013 GSS, seniors aged 65 years and older without close friends were less likely to positively rate their physical health. In particular, 34% of seniors who had no close friends stated that their state of health was excellent or very good, compared with 54% of those who reported having three or more close friends.

The proportion of people who reported having no close friends also varied by province, with the lowest proportion in Prince Edward Island (2%) and the highest in Quebec (8%). In Quebec, the proportion of seniors aged 75 and older that had no close friends was significantly higher than in all other provinces. In 2013, this proportion was 28%, compared to 11% for all other provinces combined.

Proportion of people with three or more friends on the rise

Data from the 2013 GSS are consistent with previous findings and show that a larger number of close friends is linked to better self-rated health and greater life satisfaction (Sinha 2014).

Although the proportion of people having no close friends has remained relatively stable since 2003, there has been an increase in the proportion of people with a greater number of friends (Chart 1).

Description for Chart 1

In 2013, 75% of Canadians reported having three or more close friends, compared with 70% a decade earlier (Table 1). This increase was reported by both men and women, and for most age groups. Youths aged 15 to 24 years were the most likely to have at least three close friends (88%). One of the reasons for the greater number of friends among young people is that, in any given month, they are more likely to meet new people with whom they plan to remain in contact (Sinha 2014).

In general however, proportionally, more Canadians reported meeting new people with plans of staying in contact with them in 2013 than 10 years earlier. This was true for all age groups, but especially among the youngest respondents (Chart 2).

Description for Chart 2

Social contact with acquaintances or other friends

Being able to rely on numerous acquaintances or ‘other friends’ to provide access to varied resources is at the heart of the social capital concept, based on the individualist approach (Lin 2000).In 2013, 7% of Canadians stated that they had no other friends or acquaintances, a slightly higher proportion than in 2003 (6%) (Chart 3). Despite this slight increase in the proportion of Canadians with no other friends or acquaintances, there was still an increase in the percentage of people with a high number of other friends or acquaintances (Chart 3).

Description for Chart 3

Thus, the proportion of people with more than 10 other friends was up from 56% in 2003 to 60% in 2013 (Table 2). However, this increase occurred mainly among the youngest Canadians. For example, the percentage of Canadians aged 25 to 34 who had more than 10 ‘other friends’ increased from 58% in 2003 to 70% in 2013. In contrast, among Canadians aged 55 and older, this percentage remained stable.

The marked increase in the use of virtual social networks, such as Facebook and Twitter, especially by youth, may be the source of this growing gap between age groups. While almost all youth aged 15 to 24 used social networking sites (96%), this was the case for 36% of seniors aged 65 and older.

When considering only users of social networks, the proportion of persons aged 15 to 24 with more than 10 other friends was almost the same as the proportion for all persons in this age group (77% and 75% respectively). However, among seniors aged 75 and older, who used such sites, the proportion of those with more than 10 friends was higher (58% compared with 43% for all seniors in this age group) (Chart 4).

Description for Chart 4

Exchanging favours with neighbours more frequent

To some extent, neighbours fall into the category of ‘weak ties’. However, given their close physical proximity, neighbourhood contacts also encourage the sharing of certain types of unique resources, such as exchanging favours, sociability, development of trust, and a safe environment for children.

In 2013, slightly more than 4 in 10 Canadians reported knowing many or most of their neighbours, a proportion that is unchanged from 2003 (Table 3). Despite this stability in the number of known neighbours, favours were being done more frequently than 10 years earlier. The percentage of Canadians who reported having done a favour for a neighbour in the past month rose from 61% in 2003 to 70% in 2013 (Table 4). In addition, the proportion of those who said that a neighbour had done them a favour was also up over the same period.

The likelihood of knowing neighbours and exchanging various favours varied by age group and province. Seniors aged 65 to 74 (51%) and those 75 and older (47%) were most likely to know many or most of their neighbours. In comparison, this was the case for only 29% of young adults aged 25 to 34. This finding is mainly due to the fact that older persons are more likely to have lived in their city or local community longer. For example, while 85% of seniors said they have lived in their city or local community for 10 or more years, this was true for only 46% of young adults aged 25 to 34.

There were also somewhat significant differences between provinces in terms of the likelihood of knowing neighbours. In 2013, 37% of the residents of Alberta reported knowing many or most of them, compared with 68% of people in Newfoundland and Labrador. Once again, residential mobility helped to explain the differences. For example, in 2013, Newfoundland and Labrador had the highest proportion of people reporting having lived in their city or local community for 10 or more years (78%).

The place of residence also mattered. People who lived in a census metropolitan area (CMA) were less likely to know many or most of their neighbours (38%) than those who lived outside a CMA (67%).

However, residential stability does not explain everything. People with broader social networks were also more likely to know their neighbors than those who were more socially isolated (Chart 5). For example, in 2013, 36% of people who had 1 or 2 close friends reported knowing many or most of their neighbours, compared with 48% of those who reported having 6 to 10 close friends. The same type of association existed with respect to ‘other friends’ and acquaintances.

Description for Chart 5

Decline in the frequency of social contact in person with friends

More frequent contact with friends can strengthen bonds, feelings of reciprocity and belonging to the group. Despite a greater number of Canadians reporting having more close friends and other friends, they see each other less often than was the case 10 years earlier. In particular, the proportion of people who reported having seen their friends a few times or more a week fell 12 percentage points from 56% in 2003 to 44% in 2013 (Table 5).

People who have a larger number of friends, including close friends and other friends, have a greater tendency to see them more often. However, the decrease in the frequency of contacts in person with friends held true for those with a more limited number of close friends as well as those with three or more close friends (Chart 6).

Description for Chart 6

There were similar trends, albeit less pronounced, in communication with friends by telephone, Internet or email (excluding texting). In 2003, 60% of Canadians had communicated with their friends a few times or more a week, a proportion that had dropped to 54% in 2013 (Table 5).

Despite this decrease in the frequency of contacts with friends, the vast majority of people were satisfied with the frequency of the contacts (Sinha 2014). It is possible that more recent ways to communicate with friends and relatives, such as text messages, partly explain the decrease in the frequency of contacts. In 2013, 57% of Canadians reported communicating by text message with their friends. Data on this type of contact were not collected in 2003 and 2008.

It should be noted that Canadians did not appear to have replaced in-person contacts and communication by telephone or Internet with their friends with more frequent contact with relatives (excluding the people they live with). Indeed, Canadians were less likely to see relatives a few times or more a week than 10 years ago (26% in 2013 compared with 38% in 2003) (Table 6). They were also less inclined to communicate frequently with them than they were a decade ago. This decreased frequency of contact with relatives was reported by all age groups.

Increase in contacts with friends from an ethnic group visibly different from their own

In general, people tend to choose friends with socio‑demographic, economic and cultural characteristics similar to their own (Fischer 1982; McPherson et al. 2001). One of the factors explaining the relative ethnic homogeneity of social networks is the spatial distribution of the population in Canadian neighbourhoods and regions. Outside major population centres like Toronto and Vancouver, the chance of coming into contact with people of varied origins is less common.

That being said, these opportunities may be increasing with the socio‑cultural diversification and growth of the population living in census metropolitan areas. This seems to be the case, based on data from the GSS: the proportion of people in contact with at least a few people from an ethnic group visibly different from their own has increased since 2003.^Note 7 In contrast, the percentage of Canadians who were not in contact with friends from an ethnic group visibly different from their own was down (Chart 7).

Specifically, in 2013, 59% of Canadians stated that at least a few of their friends they had contact with in the past month came from an ethnic group visibly different from their own. In comparison, this proportion was 54% in 2003 (Chart 7).

Description for Chart 7

Young people were much more likely to have had contact, in the past month, with friends from an ethnic group visibly different from their own. In 2013, 77% of people aged 15 to 24 had such contacts, compared with 43% of persons aged 65 to 74 and 39% of seniors 75 and older. These differences are due mainly to the fact that the proportion of people belonging to a visible minority is lower among persons aged 65 and older.^Note 8

The upward trend between 2003 and 2013 in the likelihood of having contact with friends from other groups was also most notable among young Canadians. Among people aged 15 to 34, for example, the proportion who had contact with friends from an ethnic group visibly different from their own increased from 65% to 72% between 2003 and 2013. In comparison, between 2003 and 2013, there was little change in the percentage of people aged 55 to 74 with such contact (Chart 8).

Description for Chart 8

Despite this increase in contacts with visibly different people, the likelihood of having contacts with friends with a different mother tongue remained mostly unchanged over the decade.

In 2013, 6 in 10 Canadians only had contact, in the past month, with friends who shared the same mother tongue, the same proportion as in 2003. This proportion varied widely by province, reflecting the regional distribution of official language and non-official language populations in Canada. In 2013, Ontario recorded the smallest proportion (54%) of people having contact only with friends sharing the same mother tongue. Newfoundland and Labrador had the highest proportion, at 82% (Table 8).

Lastly, the proportion of people aged 15 and older who reported that all of their friends were of the same sex as them fell slightly from 20% to 18% between 2003 and 2013 (Table 9). The decrease was much more significant among those aged 25 to 34, compared to any other age group.

Factors associated with social networks

In addition to differences by sex, age groups and provinces, a number of other factors were linked to the size and diversity of social networks, as well as the frequency of contact within these networks. In general, people with higher levels of education and income, and those who are employed have larger and more extensive networks (Moore 1990). Various personality traits, notably extroversion and agreeableness, can also impact individuals’ social networks (Ozer and Benet-Martinez 2006). Recent immigrants may have less well-developed social networks, especially in terms of weak ties, which can make their path to employment more difficult (Thomas 2011).

In recent years, research has examined the impact of social networking sites on the nature of interpersonal relations (e.g., Hampton and Wellman 2003; Ellison, Steinfield and Lampe 2007). How do frequent users of networking sites compare to non-users?

After eliminating the effects of individual characteristics, such as age, level of education and sex, it was found that Canadians who frequently used social networking sites were more likely to have three or more close friends (predicted probability of 0.82 compared with 0.71 for non-users).^Note 9 They were also more likely to have a few other friends or acquaintances and less likely to have no close friends (Table 10). These findings are not surprising, since the nature and purpose of these sites are to maintain social contacts with many people. But what about in-person contact with these ‘friends’, who may sometimes be only virtual?

According to certain studies, people who are more active on social networks also have more frequent non-virtual relationships with members of their networks (e.g., Young 2011; Johnston et al. 2010). Findings from the GSS are consistent with these studies, as users of social networking sites were more likely than non-users to see their friends in-person a few times or more a week (Table 11).

The Internet is sometimes described as a window to the world. In keeping with this perception, social networking site users were more likely to have been in contact with at least a few people from an ethnic group visibly different from their own in the past month (predicted probability of 0.64 compared with 0.52 for those who never used social networking sites).

Watching television does not hinder social contact

According to Putnam, part of the decline in the level of social capital in the United States is due to the increased time spent watching television alone as a leisure activity, at the expense of group or community activities that facilitate building links and creating social capital (Putnam 2000).

However, in 2013, the proportion of people who had no close friends was higher among those who do not usually watch television (Table 10). Specifically, the probability of people who did not watch television having no close friends was twice as high as that of people who watched 10 to 19 hours of television per week (probabilities of 0.07 and 0.03 respectively).

Furthermore, there was no association between the habit of watching television and the probability of having a higher number of close friends (three or more) or other friends or acquaintances (more than 10). Some people may have replaced television watching by Internet use.

People who live in the same city for 10 or more years have more social contacts

Despite the growing popularity of ‘virtual friendships’, data from the 2013 GSS show that the spatial proximity between individuals remains a key factor in the nature and form of social contacts with friends.

People who resided in their local community for long periods of time were slightly more likely to report having at least three close friends, and more than 10 other friends (Table 10). They even tended to see and communicate with their friends more often, when other individual factors were taken into account (Table 11). These results show that geographic mobility, although it may result in economic gains, may lead to losses in social capital.

Living in a census metropolitan area (CMA) may also impact social networks. People living outside a CMA were almost twice as likely to know many or most of their neighbours, compared to people who lived in a CMA (67% versus 38%). However, the opposite was true in terms of the ethnic diversity of social networks: the probability of having contacts with friends from a visibly different ethnic group was higher for residents of CMAs (Table 11).

Section 2: Trend in civic engagement and volunteerism

The book, Bowling Alone, released 15 years ago, illustrated how social capital, notably trust in one another, had declined significantly in the United States since the 1960s and 1970s. According to Putnam, one of the key elements in that decline was the decrease in civic engagement, that is, the participation of Americans in organizations, associations or groups.^Note 10 What is the situation in Canada?

Although it is not possible to compare the current situation to that of the 1960s and 1970s, it is possible to examine the trends, over the past decade, in the level of Canadians’ civic engagement.

In 2013, about two-thirds of Canadians participated in or were member of a group, an organization or an association (65 %), compared with 61 % in 2003 (Table 12). However, monthly participation in group activities remained unchanged during the period. In 2013, 48% of people had participated in group activities or meetings at least once a month, the same proportion as in 2003.

Interestingly, there was even an increase in the participation in group activities and meetings among seniors aged 75 and older. In this age group, the monthly participation rate rose from 36% to 48% between 2003 and 2013, likely reflecting the better health status of older seniors.^Note 11

In 2013, Quebec (36%) and New Brunswick (46%) were the two provinces with the lowest rates of monthly participation in group activities. Among other provinces, the participation rate varied between 51% and 55%.

Various studies have shown that participation in organizations varied quite significantly from country to country based on different contextual and cultural factors (Curtis et al. 2001; Bjørnskov 2006). When Canadians are compared based on their region of birth, there are relatively significant contrasts in terms of monthly participation in group activities and meetings (Chart 9).People born in the United States had some of the highest monthly participation rates, a finding in keeping with comparative studies that also showed above average engagement by Americans (e.g., Curtis et al. 2001).

Description for Chart 9

Civic engagement and diversification of social networks

People who participate in group activities or attend meetings can expand their networks of strong or weak ties, notably by becoming friends with people whose cultural and demographic characteristics are different from their own. According to the data from the General Social Survey, this would appear to be the case, at least to some degree.

In 2013, among people who participated in at least one group, organization or association, almost 3 in 4 (73%) had met at least a few people there who were part of an ethnic group visibly different from their own. This proportion was up slightly from 2003 (70%).

Participants in a youth group (83%), union or professional association (79%) and political organization (78%) were the most likely to have met a person from an ethnic group visibly different from theirs.

This proportion was lower among people who participated in a seniors’ group (57%) or social club, such as the Kiwanis Club, Knights of Columbus or the Legion. These results reflect those presented in Table 7, which show that young people are more likely than seniors to have ethnically-diverse social networks.

The sex of the people met while participating in group activities varied by the type of organization. For example, 12% of members or participants in a union reported that all the people they met were of the same sex. In comparison, this was the case for 20% of members or participants in a sports or recreational organization, and for 21% of those in a social club.

Volunteer participation remains stable, though volunteers devote fewer hours

Involvement in groups and associations can be motivated by a variety of factors: interest in playing a sport, a cultural activity, sociability and so on. The desire to contribute to a cause, by giving of one’s time to a charitable or not-for-profit organization, may also lead people to participate in different group activities. Volunteerism is often considered an act potentially contributing to social capital, because it enables people from different backgrounds to work together toward a common goal (Wu 2011). It should be noted that, in this section, the results were drawn from the 2013 General Social Survey - Giving, Volunteering and Participating, and from the 2004, 2007 and 2010 Canada Survey of Giving, Volunteering and Participating.

As was the case for monthly participation in group activities or meetings, volunteering has remained stable over the past decade. In 2013, 44% of people aged 15 and older were involved in some form of volunteer work, compared with 45% in 2004. However, compared with 2004, the average number of hours devoted to volunteer activities was down (see Turcotte 2015 for more details on the results of the 2013 General Social Survey - Giving, Volunteering and Participating).

Section 3: Trends in feelings of trust

Trust in others is a central and fundamental element of social capital, as defined by the collective approach. If trust between close friends and relatives is important to the survival of networks with strong ties, then trust in strangers would have considerably farther reaching impacts for the whole community.

Generalized trust is often considered an element facilitating social contacts: higher levels of trust means lower transaction costs and improved likelihood of productive interactions. One of the reasons for the recognized importance of generalized trust is its positive impact on a number of government policy objectives^Note 12.

The classic question used to measure generalized trust, and the question that has already been used in numerous studies on this topic, is “generally speaking, would you say that most people can be trusted or that you cannot be too careful in dealing with people?”

In response to this question, more than half of Canadians reported that most people can be trusted (54% in 2013, which is virtually unchanged from 2003) (Table 13). Presented differently, 46% of Canadians felt that you cannot be too careful in dealing with people.

Young people aged 15 to 24, even if they had more friends than average, were the least likely to think that most people can be trusted (48% in 2013). According to Putnam, a key factor in the decline in generalized trust is generational replacement, that is, the fact that younger generations of Americans tend to be less engaged socially and to trust others less (Putnam 2000). However, it is impossible to verify whether Canadian baby boomers, when they themselves were 15 to 24 years old, expressed a higher level of generalized trust than members of the new generations.

Quebec (36%), and to a lesser degree New Brunswick (51%), recorded the lowest levels of generalized trust. These were also the two provinces with the lowest monthly participation rates in group activities and meetings. Other studies have also found that generalized trust was lower in Quebec than in other provinces, though no definitive explanations for this difference have been offered (Kazemipur 2006).

Comparative international studies have shown that the level of generalized trust varied widely from country to country, suggesting that trust may be linked to more general social and cultural characteristics (Park and Subramanian 2012). Significant differences are also found when examining generalized trust based on individuals’ region of birth.

In 2013, the percentage of people who stated that most people can be trusted was highest among people born in Northern Europe and the United States (68% in both cases). In contrast, the lowest proportions were among people born in the Antilles and Bermuda (41%) and in Central Africa (34%) (Chart 10).^Note 13

Description for Chart 10

Trust in neighbours and strangers

People may have different reference points in mind when asked if ‘most people’ can be trusted. Consequently, it is possible that some people’s levels of trust change when a specific reference to people in the neighbourhood or to strangers is added.

The 2013 GSS asked respondents “using a scale of 1 to 5, where 1 means ‘Cannot be trusted at all’ and 5 means ‘Can be trusted a lot’, how much do you trust each of the following groups of people”. In this case, neighbours and strangers are the focus.

In 2013, the average levels of trust of Canadians toward people in their neighbourhood, using this scale of 1 to 5, was 3.7 (Table 14). This score was unchanged from 2003. The average level of trust toward strangers was lower, but had improved from a decade ago (2.4 in 2013 compared with 2.2 in 2003).

Older Canadians expressed a higher level of trust in their neighbours and strangers than younger Canadians. For example, persons aged 75 and older gave a trust score of 4.1 out of 5 for people in their neighbourhood, compared with 3.4 out of 5 for people aged 15 to 24. This result is especially interesting considering that older Canadians are generally more likely to indicate that they do not feel very safe in terms of crime rates in their neighbourhood (Fitzgerald 2008).

Provincially, Quebec had the lowest trust scores both in terms of neighbours (3.5 out of 5) and strangers (2.1 out of 5). Quebeckers were also most likely to consider that one can never be too careful in dealing with people (64% compared with 43% of Ontarians).

Almost half of Canadians believe that their neighbour would return their lost wallet or purse

To give more concrete meaning to the classic generalized trust question, instruments were developed to measure trust in others in more specific contexts. In the GSS, respondents were asked “if you lost a wallet or purse that contained two hundred dollars, how likely is it to be returned with the money in it, if it was found by someone who lives close by?”

In 2013, 45% of people said that it would be very likely that someone would return their wallet or purse, the same proportion as in 2003 (Table 15). Canadians aged 75 and older were proportionally more likely (60%) to believe they would recover their wallet or purse, compared to people aged 35 to 44 (48%). Young Canadians were the most sceptical of their neighbours: about one third (32%) believed they would recover their lost money if found by someone living close by.

At the provincial level, residents of Newfoundland and Labrador, who were most likely to know their neighbours, were also most likely to believe their wallet or purse would be recovered (63% compared with 44% in Ontario and 42% in Quebec).

Factors associated with generalized trust

Additional regression models were developed to better understand the relative importance of the different factors associated with trust. One model addressed generalized trust and the other examined the likelihood that a lost wallet would be returned by a neighbour. These models took into account demographic characteristics (such as age and sex) and socio-economic characteristics (such as level of education, immigrant status and province of residence). The models also take into account the participation in group activities or meetings, the size of social networks of friends and acquaintances, and the use of social networking sites.^Note 14

Table 16 shows that when other characteristics are held constant, older Canadians remained more likely to say that most people can be trusted and to believe that they would recover a lost wallet or purse if found by a neighbour. For example, the probability that seniors aged 75 and older believed that they would recover their lost wallet or purse was almost 0.60. This was 1.7 times higher than the probabiliy for people aged 15 to 34.

Civic engagement is considered a central factor in developing generalized trust (Putnam 2000). Various studies have illustrated this relationship, showing that people who participated in groups, associations or organizations expressed a higher level of trust in others (e.g., Paxton 2007). In keeping with the findings of these studies, Canadians who participated in group activities or meetings were more likely to report that most people can be trusted (Table 16).

There was also a close relationship between the size of social networks and generalized trust, as has been noted in other earlier studies (Helliwell and Wang 2010). For example, the probability of trust in most people rose from 0.43 for people with no close friends to 0.59 among those with between 6 and 10. The same type of association existed with respect to ‘other friends’ and acquaintances. However, believing that you can “never be too careful dealing with people” can reduce the propensity of individuals to create bonds and make new friends.

Lastly, life experiences, whether positive or negative, can change people’s perceptions of their fellow citizens. As part of the GSS, respondents were asked if, in the past five years, they had experienced discrimination because of their sex, ethnicity or culture, race or colour, physical appearance, religion, sexual orientation, age, disability, language or some other reason. People who responded affirmatively to one of these questions were deemed to have experienced discrimination.

Overall, experiencing discrimination was a factor in reducing social trust. While holding other individual characteristics constant, people who had experienced discrimination in the past five years had a lower probability of believing that most people can be trusted (0.46 ), compared to those who had not experienced discrimination (0.58).

Conclusion

The concept of social capital, which in the past was an academic concept used by sociologists, has become part of public debate (OECD 2014). The value of social capital from a policy perspective is based on the idea that social relationships of trust and reciprocity, beyond the personal satisfaction gained from them by individuals, have a positive impact on society as a whole (Scrivens and Smith 2013).

Based on the GSS, Canadians had more extensive social networks of close friends (strong ties) and other friends and acquaintances (weak ties) than 10 years earlier. The popularity of social networking sites has likely played a role in the growth of social networks, as those who frequently used social networking sites were more likely than non-users to have a larger number of friends.

Despite this increase in the number of friends, Canadians saw their friends somewhat less frequently in 2013 than in 2003. They were also less likely to communicate with them by telephone or email. It is possible that new methods of communication, such as text messages, which were not widely used a decade ago, have partly replaced these more traditional methods of staying in touch.

In terms of participation in organizations, associations or groups, fewer changes were reported over the decade. The participation rate in groups increased slightly between 2003 and 2013. However, the proportion of Canadians who participated at least monthly in group activities or meetings remained virtually the same during the period. The proportion of Canadians who volunteered their time also remained practically unchanged.

The same can be said of the feeling of trust. In 2013, 54% of Canadians reported that, in general, most people can be trusted, while on the other hand, 46% believed that they could never be too careful in their dealings with others. These proportions were essentially the same in 2003. Other indicators of trust in neighbours or strangers also remained virtually unchanged over the period.

In summary, the GSS data illustrate an increase in social capital, as defined by the individualist perspective (number of friends, exchange of favours between neighbours and diversity of social networks). On the other hand, the GSS shows a greater stability with respect to social capital indicators associated with the collective approach (participation in group activities, volunteering and trust).

Data sources

This report is based on data from the 2003, 2008 and 2013 General Social Survey. The target population consisted of persons aged 15 and older living in Canada’s 10 provinces, excluding people living full-time in institutions. The number of respondents was 24,951 in 2003, 20,401 in 2008 and 27,695 in 2013, for a total of 73,047 persons.

For more information on the data sources, please consult the following documents:

2013 GSS:
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SurvId=134876&InstaId=139605&SDDS=5024

2008 GSS:
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SurvId=44601&InstaId=30687&SDDS=5024

2003 GSS:
http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SurvId=1390&InstaId=5509&SDDS=5024

In this report, missing responses have been excluded from the denominator to facilitate comparisons over time. For this reason, some of the results may differ slightly from those presented in a previous report (Sinha 2014).

References

Berkman, L. 1995. “The role of social relations in health promotion”, Psychosomatic Medicine, vol. 57.

Berkman, L. and T. Glass. 2000. “Social integration, social networks, social support and health” Social Epidemiology, In L. Berkman and I. Kawach (Eds), Oxford University Press, Paris.

Bjørnskov, C. 2006. “Determinants of generalized trust: a cross-country comparison”, Public Choice, vol. 130.

Bourdieu, P. 1986. “The forms of Capital”, Handbook of Theory and Research for the Sociology of Education, published under the direction of J.G. Richardson,Greenwood Press, New York.

Boxman, E., P. De Grant and H.D. Flap. 1991.“The impact of social and human capital on the income attainment of Dutch managers”, Social Networks, vol. 13.

Chiswick, B.R. and P.W. Miller. 1996. “Ethnic networks and language proficiency among immigrants”, Journal of Population Economics, vol. 9.

Coleman, J. 1988. “Social capital in the creation of human capital”, American Journal of Sociology, vol. 94.

Curtis, J.E., D.E. Baer and E.G., Grabb. 2001. “Nations of joiners : Explaining voluntary association membership in democratic societies”, American Sociological Review, vol. 66.

Dekker, P. 2004. “Social capital of individuals: Relational asset or personal quality?”, Investigating Social Capital – Comparative Perspectives on Civil society, Participation and Governance, In S. Prakash and P. Seel (Eds), Sage Publications, Thousand Oaks.

Ellison N. B., C. Steinfield and C. Lampe. 2007. “The benefits of Facebook ‘friends’: Social capital and college students’ use of online social network sites”, Journal of Computer-Mediated Communication, vol. 12.

Erickson, B.H. 2001. “Good networks and good jobs: the value of social capital to employers and employees”, Social Capital – Theory and Research, In N. Lin, K. Cook and R.S. Burt (Eds), Transaction Publishers, New Jersey.

Fischer, C.S. 1982. To Dwell among Friends: Personal Networks in Town and City, University of Chicago Press.

Fitzgerald, R. 2008. Fear of crime and the neighbourhood context in Canadian cities, Crime and Justice Research Paper Series. Statistics Canada catalogue no. 85-561.

Fukuyama, F. 1996. Trust: The Social Virtues and the Creation of Prosperity, Free Press Paperbacks, New York.

Hampton, K. and B. Wellman. 2003. “Neighboring in Netville: How the Internet supports community and social capital in a wired suburb”, City and community, vol. 2, no. 4.

Helliwell, J.F. and S. Wang. 2010. Trust and Well-Being, working document no. 15911 of the National Bureau of Economic Research, Cambridge.

Granovetter, M. 1973. “The strength of weak ties”, American Journal of Sociology, vol. 78.

Johnston, K., M. Tanner, N. Lalla and D. Kawalski. 2010. “Social capital: The benefit of Facebook ‘friends’”, Behaviour & Information Technology, vol. 32, no. 1.

Kazemipur, A. 2006. “A Canadian exceptionalism? Trust and diversity in Canadian cities”, Journal of International Migration and Integration, vol. 7, no. 2.

Kawachi, I, B.P. Kennedy and R. Glass. 1999. “Social capital and self-rated health: a contextual analysis”, American Journal of Public Health, vol. 89, no. 8.

Kennedy, B.P., I. Kawachi, D. Prothrow-Stith, K. Lochner and V. Gupta. 1999. “Social capital, income inequality, and firearm violent crime”, Social science and medicine, vol. 48.

Lin, N. 2001. Social Capital: A Theory of Social Structure and Action, Cambridge University Press, Cambridge.

Luo, Y., L.C. Hawkley, L.J. Waite and J.T. Cacioppo. 2012. “Loneliness, health and mortality in old age: A national longitudinal study”, Social Science and Medicine, vol. 74.

McPherson, M., L. Smith-Lovin and J.M. Cook. 2001. “Birds of a Feather: Homophily in social networks”, Annual Review of Sociology, vol. 27.

Moore, G. 1990. “Structural determinants of men’s and women’s personal networks”, American Sociological Review, vol. 55.

OECD. 2014. How’s Life? 2013: Measuring Well-being, OECD publishing, Paris.

Ozer, D., and V. Benet-Martinez. 2006. “Personality and the prediction of consequential outcomes”, Annual Review of Psychology, vol. 57.

Park, C, and S.V. Subramanian. 2012. “Voluntary association membership and social cleavages: A micro-macro link in generalized trust”, Social Forces, vol. 90,no. 1.

Paxton, Pamela. 2007. “Association membership and generalized trust: A multilevel model across 31 countries”, Social Forces, vol. 86.

Perissinotto, C.M., I. Stijacic Cenzer and K.E. Covinski. 2012. “Loneliness in older persons: A predictor of functional decline and death”, Archives of Internal Medicine, vol. 172, no. 14.

Putnam, R. 1993. Making Democracy Work: Civic Traditions in Modern Italy, Princeton University Press, New Jersey.

Putnam. R. 2000. Bowling Alone: The Collapse and Revival of American Community, Simon and Schuster, New York.

Sampson, R.J., Raudenbush, S.W. and E. Earls. 1997. “Community structure and violent crime: A multilevel study of collective efficacy”, Science, vol. 277.

Scrivens, K and C. Smith. 2013. Four Interpretations of Social Capital: An Agenda for Measurement, OECD, Paris.

Sherbourne, C., R. Hayes and K. Wells. 1995. “Personal and psychological risk factors for physical and mental health outcomes and course depression amongst depressed patients”, Journal of Consulting and Clinical Psychology, vol. 63, no. 3.

Shields, M. 2008. “Community belonging and self-perceived health”, Health Reports, vol. 19, no. 2.

Sinha, M. 2014. “Canadians’ connections with family and friends”, Spotlight on Canadians: results from the General social survey, Statistics Canada catalogue no. 89-652.

Thomas. 2011. “Personal networks and the economic adjustment of immigrants”, Canadian Social Trends, Statistics Canada catalogue no. 11-008.

Turcotte, Martin. 2015. “Volunteering and charitable giving in Canada”, Spotlight on Canadians: results from the General social survey, Statistics Canada catalogue no. 89-652.

Verba, S., K. L. Schlozman and H.E. Brady. 1995. Voice and Equality: Civic Voluntarism in American Politics, Harvard University Press, Cambridge.

Wellman, B., and S. Wortley. 1990. “Different strokes from different folks: Community ties and social supports”, American Journal of Sociology, vol. 96, no. 3.

Wu, Huiting. 2011. Social Impact of Volunteerism, Points of Light Institute.

Young, Kirsty. 2011. “Social ties, social networks and the Facebook experience”, International Journal of Emerging Technologies and Society, vol. 9, no. 1.

Notes