Aging of the Orchestra Audience and the Fallacy of “Demographic Destiny”

Like Disney's Chicken Little, the Chicken Little Think Tank would have us believe the "Sky is Falling" in Classical Music Culture!
Like Disney’s Chicken Little, the Chicken Little Think Tank would have us believe the “Sky is Falling” on Classical Music Culture!

In a February 2011 NEA Research Report, “Age and Arts Participation: A Case against Demographic Destiny” by Mark J. Stern, we find a refutation of the so-called dire data that is what the author is calling the “Demographic Destiny” of the graying of arts audience.  The description from the link to the reports above:

Mark Stern, University of Pennsylvania, analyzes the relationship between age and arts participation in the Survey of Public Participation in the Arts data for 1982, 1992, 2002, and 2008. The report concludes that age and year of birth are poor predictors of arts participation and that the age distribution of art-goers now generally mirrors that of the U.S. adult population.

Which is pretty much both Matthew Guerrieri and I surmised given a more nuanced look at the data.  Basically, the aging audiences is simply a function of Demographic Evolution rather than Demographic Destiny.

The preface from the piece describes studies commissioned by the NEA by four researchers or teams of researchers using the 2008 SPPA (Survey of Public Participation in the Arts) data.  One was published in 2010 while the other three, including Stern’s, were published in 2011.

Working along quite different lines, Mark Stern similarly concludes that arts education is the most important known factor in influencing arts participation trends. But he is much more skeptical about the impact of other variables, especially age. Practically since the SPPA began, in 1982, there has been much talk about the “graying” of arts audiences. And while it is certainly true that the audiences for many art forms tracked by the SPPA are aging more rapidly than the U.S. population, Stern brings out the sobering fact that age and generational cohort differences account for less than 1 percent of the variance in the total number of arts events that Americans attended over the period of 1982–2008. Observing that arts attendance may be far less dependent on age than usually considered, he gives the lie to the notion of “demographic destiny” when it comes to arts engagement.

While it may be pointed out that the data deals with arts engagement or participation, but since the aging of arts audiences issue is derived from the same dataset this criticism is a non-issue.  In a previous post in this series looking at Aging Orchestra Audiences I listed a variety of ways that we tend to misinterpret data, or how data and studies interpreting them may be biased especially as it pertains to publication, but I didn’t really discuss the issues with what the SPPA surveys actually represent.

The data in the SPPA surveys differ from many of the Orchestra surveys in that they may be more prone to Self-Reporting bias.  The data was collected via in-person or phone interviews and subjects were asked about attendance and participation over the previous year(s).  Most of the Orchestra surveys were actually given at events (e.g. a survey in the program of a concert at which an audience member is actually in attendance).  Self-report bias is well documented and studied in the clinical and epidemiological literature which often shows a bias towards what is considered “Socially Desirable” responses.

Since the claim that, say, Classical Music as a cultural phenomenon is out of touch with youthful audiences (hence the “graying of Classical Music audiences” issue) we’re left with the ironic situation that in such large scale Self-Report surveys as the SPPA, given the natural Social Desirability Bias for self report we might wonder which cohorts are the most likely to report attendance and engagement.  If younger audiences feel as if the arts are out of touch with them, then perhaps they would be far less likely than “elderly” respoders to report attendance, engagement in the arts or the desire for either.  This would obviously skew the audience age upwards.

The predictive powers of age and cohort were never particularly strong, and they declined over time.

The above being said, Mark Stern, in his report discusses other ways the data can be skewed due to self-reporting.  His introduction starts with “Age Consciousness”:

A century ago, many Americans did not know exactly how old they were, so they often would round off their age to the nearest five years, a phenomenon that demographers call “age-heaping.” As late as 1910, for example, the U.S. Census listed 24 percent more 20-year-olds than 19-year-olds. The disappearance of age-heaping and its replacement by age consciousness — an increased sensitivity to the role of age and generation on behavior — by the middle of the 20th century represents a profound change in how Americans thought about their lives and their relationship to the rest of society. Today, age consciousness has so penetrated our society that one’s membership in a particular generation or birth cohort is often offered to explain a variety of behaviors — from consumption decisions to political preferences.

Age consciousness has affected our understanding of arts participation as well. Differences in rates of arts participation of the Baby Boomers and earlier and later generations have been of particular concern. Yet we might ask: has age consciousness gotten out of hand?

Has it gotten out of hand?  Likely.  As I mentioned in my post about types of biases, Philip Tetlock has shown that “Hedgehogs” who tend to have singular worldviews and who tend to be the most vocal about doom and gloom also tend to be the most wrong in their predictions.  Stern, taking a more nuanced view of the data, is more like Tetlock’s “Foxes” who are far more cautious and use a variety of tools rather than a simple metric (such as a “median age of audiences which is rising faster than the median age of the population to show the decline of Classical Music”) to inform his worldview.

Whether the view is that “Classical Music is dying” or “Classical Music is perfectly healthy”–both are entrenched viewpoints, and more than likely to be wrong in some fundamental way.  Douglas Dempster, made a similar statement about the issue of declining interest in Classical Music over ten years ago (bolded part my emphasis):

I haven’t offered anything approximating an exhaustive survey of the known data on the classical music audience. But the studies reviewed here make it perfectly clear that critics have, perhaps in a spate of millennial fever, greatly exaggerated the demise of classical music at the end of the 20th century. Even worse, however, they have witnessed very complex trends in the culture of classical music and reduced them to the morally simplistic calculus of “rise” and “decline.” Musical and cultural critics misinterpret economic, demographic, and technological changes affecting the world of classical music as signaling some spiritual decay in the culture of classical music itself. The audience for classical music is not withering, but technological, sociological, and economic forces are reshaping that audience in important ways.

Here is what Stern says about the rising age of audiences:

Much of the shift of audience toward older age groups resulted from the aging of the population. If we correct for this factor, then audience-share figures show a less dramatic aging pattern. For example, the decline in the participation index for young adults was only from 28 to 25 percent, using corrected figures, while the increase in the older middle-age audience was less than one percent (23.6 to 24.3 percent). The increase in the older-adult audience remained a substantial 5 percentage points.

Overall, audience shares in 2008 more closely tracked the distribution of the entire population in 2008 than it had earlier. While the aging of the arts audience was real, it was less a product of changes in people’s tastes and behavior, than of the aging of the overall population. With a few exceptions — especially jazz and ballet — changes in audience share by age groups for arts activities tracked changes in the composition of the population.  pg. 18

Again, aging audiences are likely just a function of Demographic Evolution.  Read the rest of Stern’s piece for a more sobering viewpoint of the aging of arts audiences.


For more post in this series, visit the Aging of Orchestra Audiences page.


  1. […] Since the 60s and 70s though we’ve been having this rise in ethnic orchestras as I’ve been blogging. More than half of these Chinese ensembles were formed in the 80s and 90s. The same can be said of the bay area Chinese Orchestras and Ensembles. As the proportion of White American population shrinks relative to the total population, I’d suggested that the so-called decline in Classical Music was simply a reflection of that changing demographic. It should be noted that the median age of the White Population in the US is also growing faster than the median age of the population of the US as a whole. What this all suggest to me is simply a A Function of Demographic Evolution and not Demographic Destiny. […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s