Cultural engagement across the life course: Examining age–period–cohort effects

Abstract

Longitudinal studies of cultural engagement are uncommon. This paper explores whether there are stable latent clusters of cultural engagement over time and whether age–period–cohort (APC) effects explain within-person changes between these clusters across the life course. Using data from the British Household Panel Survey (N = 25,149), I conduct longitudinal latent class analysis to uncover the latent patterns of cultural consumption across nine activities: cinema, watching live sport, attending the theatre, eating at a restaurant, doing DIY, attending an evening class, meeting with local groups, volunteering, and going to a club/pub. Then, I explore whether APC effects are observable in the data using scatterplot smoothing, lowess regression. Six latent clusters are consistently observed in each wave: None, DIY, Restaurant/Pub, Cinema/Theatre, Volunteer, and All (almost all activities). A large number of respondents have highly stable patterns of cultural engagement but there is also a substantial degree of variation. Transitions between latent clusters over time occur between those that are compositionally similar (in terms of activities). These changes in patterns of cultural engagement are partially explained by age effects, such as time constraints and health. Additionally, there is some evidence of cohort effects among the youngest cohorts moving into the “All” latent cluster, the cluster with the most cultural breadth.

Keywords:

• cultural engagement
• omnivore
• Bourdieu
• age
• life course
• latent classes

Acknowledgements

For helpful comments and advice, I thank Nick Allum and Malcolm Brynin; and three anonymous reviewers.

Funding

This research is funded by a grant from the Economic and Social Research Council [ES/I902341/1].

Notes

1. Ollivier (2008) uses this concept in the context of exploring cultural omnivores and does so in order to hig hlight how this openness is rhetorically organised.

2. Bourdieu (1984) defines habitus in various ways but in short it is “a structuring structure”, in that it organises the social world and classifies it, and it is “a structured structure”, in that this classificatory system is itself the product of “internalisation of the division into social classes”. The habitus is a way of relating to the world which emerges from being embedded within a system of relatively stable social expectations and norms.

3. The Dissimilarity Index is a measure of the difference between the observed and expected cell frequencies. In other words, it measures how well the model predicts the actual results observed in the data. According to Vermunt and Magidson (2005), the score indicates the proportion of observations that would need to be moved, from one cell to another, in order to get a perfect fit.