1. Introduction

View the most recent version.

A number of international studies have explored the socioeconomic composition of culture audiences. Common findings are that income, occupation and education are all associated with greater attendance at, or participation in, most culture activities (for early work in this area see Baumol and Bowen, 1966; Book and Globerman, 1975; Globerman and Book, 1977; for literature reviews, see, Ateca-Amestoy, 2005; Borgonovi, 2004; Seaman, 2005; Sturgis and Jackson, 2003). This paper contributes to that literature by using the culture supplement of Statistics Canada's 2005 General Social Survey (GSS) to investigate the extent to which socioeconomic characteristics influence the number of times Canadians aged 15 and older participate in the following activities: attend theatrical performances or popular music performances; visit historic sites, conservation areas or nature parks; go to a public gallery or art museum; listen to music; go to a movie or drive-in; watch videos; read books or magazines; and use a library during a year.¹

In so doing, we also add to other Canadian studies which have examined culture participation from a multivariate perspective. In one such study, Bourdeau (2002) found that gender, age, educational level, household income and whether one was born in or outside of Canada are all important determinants of participating in at least one of the following culture activities: dancing, acting, singing as part of a group, choir or solo, writing, taking photographs and playing musical instruments. These factors were also positively associated with participating in culture events as audience members or visiting cultural or heritage sites as visitors. In another study, Fisher and Preece (2003) showed that adult individuals who attended only "classical music" concerts (i.e., attended at least one of the following: symphonic, chamber, choral or opera music concert) were more likely to be older women and foreign-born individuals, whilst those who attended "classical music" and "other music" concerts (i.e., attended at least one pop, rock, jazz, blues, folk or country music concert) were more likely to be younger, male and urban. Fisher and Preece also noted that education and income have strong positive and significant effects on attendance at musical concerts.²

While the findings from Bourdeau (2002) and Fisher and Preece (2003) are useful, they both treat culture participation as a dichotomous phenomenon, and estimate a binary indicator variable as a function of individual socioeconomic attributes. A major disadvantage of this approach is that the importance of any demographic group at a given culture activity will be understated if that group participates an unusually high number of times per period, and be overstated in the event that the absolute frequency is unusually low (Seaman, 2005). Thus it is quite reasonable to argue that factors affecting whether or not a person participates could be significantly different from factors affecting frequency of participation. Furthermore, the focus of these multivariate studies is quite narrow, and excludes significant types of cultural consumption such as going to the cinema, reading books and magazines, viewing videos and listening to music. As Charts 1 and 2 show, these types of cultural activities constitute important forms of cultural consumption often enjoyed by Canadians of all income and age groups.³

The idea that parents' education plays an important role on cultural orientations and behaviour is well documented in international research. For instance, Borgonovi (2004), and Lewis and Seaman (2004) found that people whose mother and father possess higher levels of educational attainment are more likely to belong to culture audiences in the United States, all other things remaining equal. Other researchers such as Bergonzi and Smith (1996) and Kracman (1996) have also noted that Americans from backgrounds with a higher level of socioeconomic status are more likely to receive early exposure to culture. This in turn has an important influence on their latter consumption of most culture commodities. Despite this empirical evidence, prior Canadian quantitative research on culture participation has ignored the relationship that may exist between family background, as measured by mother and father's educational qualifications and culture participation. Instead, Canadian research has essentially focussed on the impact that standard personal characteristics have on participation behaviour.

As a complement to this international literature, as well as going beyond prior Canadian research, we provide new Canadian evidence of the influence of people's "social capital" on their level of culture consumption.⁴ However, we also go further and explore empirically the effect that an adult's relationships in later life has on his or her participation in culture activities, taking into account the educational attainment of an individual's partner or spouse. Controlling for the educational qualification of the conjugal partner is important in analyses of culture participation as it has been found to be a good representative of people's current social capital, which in turn plays a crucial role in the transmission of cultural interests and largely contributes to the formation of their preferences (DiMaggio and Useem, 1978b; Upright, 2004, Van Berkel and De Graaf, 1995).

Why focus on social correlates of culture participation in Canada? A socioeconomic profile of individuals participating in different cultural activities can provide useful insights to various stakeholders. It can give organizations concerned with the delivery and the commissioning of culture projects or programmes an understanding of those individuals who are most likely to attend their events. Also, identifying the socioeconomic and demographic groups most likely to attend cultural events can contribute to the growth of the market for those cultural events in Canada.

The remainder of the paper is organized as follows. Section 2 gives a brief overview of the 2005 Time Use Survey (subsection 2.1), presents a descriptive analysis (subsection 2.2) and investigates linear correlations (subsection 2.3). Section 3 outlines our estimation strategy (subsection 3.1) and provides a formal discussion of the variables used in our analysis (subsection 3.2). The estimates of the participation equations are presented in Section 4. Section 5 concludes. All tables are in the Appendix.

Chart 1
Participation in cultural activities by household income, Canada, 2005

Chart 2
Participation in cultural activities by age, Canada, 2005