The Rise of the Eclectic Cultural Consumer in Denmark, 1964–2004

Abstract

Existing research on cultural stratification and consumption patterns rarely presents a cross-time comparative perspective and rarely goes back before the 1980s. This article employs a unique series of surveys on cultural participation collected in Denmark over the period 1964–2004 to map the historical development of three distinct cultural consumption groups (eclectic, moderate, limited) also identified in previous research. We report two major findings. First, the eclectic (or “omnivorous”) cultural consumption group existed as far back as the 1960s and has since the 1980s comprised about 10 percent of the Danish population. Second, the major stratification variables—income, education, and social class—are strong predictors of cultural eclecticism in Denmark, and the predictive power of these stratification variables appears not to have declined in any substantive way over the past 40 years.

ACKNOWLEDGMENTS

We thank Meir Yaish, Michèle Ollivier, Torben Fridberg, and Jens Bonke for helpful comments and Guillermo Huberman for research assistance. Three reviewers made very helpful suggestions. The Danish Social Science Research Council (grant 275-07-0046) provided financial support for Jæger. Data from the 1964–1998 surveys were provided by the Danish National Centre for Social Research and data from the 2004 survey were provided by the Danish Data Archive. The 2004 survey data were collected by TNS Gallup for Trine Bille. The Archive and the original data collectors do not bear any responsibility for the results and interpretations presented in this article.

NOTES

Notes

¹ Lahire (2004, 2008) has argued that every social class is involved in dissonant cultural profiles to a certain degree, implying that omnivorousness is not exclusive to the elite.

² In the 2004 survey, it was not possible to identify managers and this class category was dropped. Judging by the distribution of the class variable, it seems that this category is captured in 2004 by the routine nonmanual category. Since we are interested in the general pattern of the class effect and not in the specific differences between categories, we do not suspect that this exception is of much significance.

³ In the LCR models presented below, models with four classes are much more complex to estimate than models with three latent classes. It is also for this reason we prefer models with three latent classes.

⁴ When interpreting Figure 2 and later figures, it should be kept in mind that the effect of the socioeconomic variables is portrayed on the x-axis (i.e., in the different income “slopes” for each cultural consumption group) and not on the y-axis (i.e., the absolute size of each group). We could easily change the predicted absolute group size by setting other fixed socioeconomic and demographic characteristics (which would not change the income “slopes”).

⁵ When we plot the probability of belonging to the limited cultural consumption group by survey year (i.e., similar to Figure 4 but with instead plotting the limited class, not shown), we find the same pattern but, not surprisingly, now reversed.

⁶ In Denmark total public social expenditure as percentage of GDP, a rough measure of the size of the welfare state, increased from around 10 percent of GDP in 1960 to around 32 percent in 2005. There are no consistent time series data for historical developments in measures of income dispersion such as the GINI index of income inequality. In 2005 the GINI coefficient for Denmark was 24 and was among the lowest in the world (by comparison, in 2007, the GINI coefficient for the United States was 45).