"In 1966, the Median Age of Orchestra Audiences was 38"
As a part of the Twentieth Century Fund survey in 1963-64, William Baumol and William Bowen published the seminal book in Cultural Economics showing their research results. The median age of arts audiences is included in their "Performing Arts: The Economic Dilemma" published by the Twentieth Century Fund in 1966.
The claim that the median age of orchestra audiences in 1966 is 38 is undecidedly false.
"In the early 1960s, the average age of the orchestra audience apparently was 38." (Sandow 2006)
"Go back to the 1930s, or the ’50s, or even 1966 (when a big study was done), and you’ll find a classical music with a median age in its thirties, not much older than the population at large." (Sandow 2011)
"Go back to 1966, and things look different. Classical music institutions didn’t have development department. The audience had a median age of 38" (Sandow 2012)
"A summary on a National Endowment for the Arts (NEA) survey states that the average age of the symphony-goer was 30 in 1937, 38 in the early 1960'2, 40 in 1982, 45 in 1992, and 49 in 2002." (attributed to Greg Sandow in Schultz 2009, pg 13)
"And in a pioneering 1966 book by William J. Baumol and William G. Bowen, Performing Arts: The Economic Dilemma, I found data from the 1960s showing that the audience in those years--in every field of the performing arts--had a median age of 38." (Sandow 2010)
There are two separate claims here. That the Median Age of Orchestras in the early 1960s (or sometimes specifically 1966) was 38. The other being that a study was done in 1966.
The Survey which Baumol and Bowen administered occurred during 1963-64. The results of the survey were published in 1966.
The Median Age of Orchestra audiences is given in the appendix as 39.
As can be seen under the "Reasonable Variants" section, Sandow is capable of stating the claim as it appears in the Baumol and Bowen book, but that still doesn't excuse misrepresentation elsewhere.