- -JT-. J-> -:-.=i<s«t;-.,»t-,..,i3¥-
1
JPIB!^^^i:'?;,T^'?v^"i* ^ " ■'->■ ' "'
Wm
iNicholas Zill and Marianne Winglee
n Foreword by Jonathan Yardley
Digitized by the Internet Archive
in 2012 with funding from
LYRASIS Members and Sloan Foundation
http://archive.org/details/whoreadsliteratuOOzill
WHO READS 1 1 1 1 R VI I Rl '
WHO READS LITERATURE?
The Future of the United States
as a Nation of Readers
Nicholas Zill
and
Marianne Winglce
Foreword by
Jonathan Yardley
SEVEN LOCKS PRESS
Cabin John, Md / Washington, DC
Who Reads Literature? is Report *22 in a series on matters of interest to the arts
community commissioned by the Research Division of the National Endowment
for the Arts.
Foreword © by The Washington Post. It originally appeared in The Washington
Post on August 28, 1989. It is reprinted here with permission of the author and
The Washington Post.
Library of Congress Cataloging-in-Publication Data
Zill, Nicholas.
Who reads literature? : the future of the United States as a nation of readers /
Nicholas Zill and Marianne Winglee : foreword by Jonathan Yardley.
p. cm. — (Research Division report / National Endowment for the Arts ; 22)
ISBN 0-932020-86-0 : $9.95
1. Books and reading — United States— Forecasting. 2. Literature-
Appreciation — United States — Forecasting. I. Winglee, Marianne. II. Title.
III. Series: Research Division report (National Endowment for the Arts. Research
Division) : 22.
Z1003.2.Z54 1990 90-33699
028.9093— dc20 CIP
Manufactured in the United States of America
Designed by Giles Bayley
Cover design by Betsy Bayley
Typeset by Bets, LTD, Ithaca, NY
Printed by McNaughton & Gunn, Saline, MI
Seven Locks Press is a Washington-based book publisher of non fiction works on
social, political and cultural issues. It takes its name from a series of lift locks
on the Chesapeake and Ohio Canal.
For more information or a catalog:
Seven I/)cks Press
P.O. Box 27
Cabin John, MD 20818
(301)320-2130
Contents
Foreword vii
Introduction 1
Chapter 1: 7
Readers of Fiction, Poetry and Drama:
How Many and Who Are They?
Chapter 2:
What the Readers Are Reading
23
Chapter 3:
Factors That Affect Literary Participation
35
Chapter 4:
Expanding the Audience:
What Can Be Done?
73
Technical Appendix
87
Notes
99
Foreword
Several years ago Walker Percy wondered aloud, in the pages of
a national magazine, about the number of Americans who actual-
ly buy, read, and discuss serious contemporary literature. It was,
he thought, a pathetically small number, perhaps one percent of
the adult population or even less: a serious literary community,
that is, of somewhere between one and two million Americans,
scarcely a consequential quantity in a culture where audiences in
the tens of millions are routinely tuned into television sitcoms and
sporting events.
This gloomy assessment caused a flurry of controversy among
the lit'ry folk; they had assumed themselves to be a far larger crowd,
if not indeed a horde, so not merely to be disabused of this notion
but to have the message brought by one of their own was anything
except pleasant. But the stir came and went quickly, in the main
because no one had any hard evidence to refute Percy's complaint
and also because, let's face it, not many outside the literary com-
munity were interested.
iiii Who Reads Literature?
Still, the questions won't go away: Who reads serious litera-
ture in the United States, and how many of them are there? These
are matters of continuing interest to authors, scholars, and others
whose careers are predicated on the existence of a viable literary
community, and they are of interest as well to those whose invest-
ment is more overtly commercial: publishers, booksellers and
wholesalers, journalists, and other media folk. Like any other group
in our economy, the literary community is a market to be identi-
fied and exploited by those with something to sell to it; thus its
size and character are matters of more than passing interest to some
persons and institutions that are not, themselves, necessarily mem-
bers of it.
In these circumstances it is useful to have a Who Reads Liter-
ature: The Future of the United States as a Nation of Readers. As
one can quickly surmise, it is the work of people who specialize
in the jargon of the social sciences: Nicholas Zill, a social psy-
chologist, and Marianne Winglee, a statistical analyst. They have
assembled a good deal of valuable information and they have
managed to make a degree of sense out of it.
As so often happens when the voodoo priests of sociology, psy-
chology and statistics work their solemn magic on human behavior,
what this survey does is tell us, in statistics and analysis, what we
know already through empirical observation: that Walker Percy was
right. Zill and Winglee have their hearts in the right place and ear-
nestly wish the evidence told them otherwise, but what their num-
bers add up to is that (a) "the proportion [of Americans] who read
serious literature of all forms in the course of a year seems to be
about 7 to 12 percent of the adult population" and that (b) "litera-
ture reading" is "stagnant or even declining, when various demo-
graphic factors indicate that it should be increasing."
Even that figure of 7 to 12 percent is shaky at best, for people
who claim to be regular readers often admit, when pressed for
specifics, that they are regular readers not of Saul Bellow and Eu-
dora Welty but of Stephen King and Danielle Steele; if we restrict
readers of literature to "those familliar with excellent but not widely
known authors, such as poets Adrienne Rich or James Merrill, then
the size of the audience for contemporary literature would become
minuscule indeed." Not merely that but— as anyone keeping a close
Foreword ix
eye on literary matters should realize— to a striking degree this
readership is defined by sex, education and income:
. . . if we had to put together a picture of a typical reader
of literature in the United States today, the survey data
indicate that the person would be a middle-aged white
female living in the suburbs of a Western or Midwestern
city. She would have a college education, and a middle-
to upper-middle class income that was not derived from
her literary activities. She would be an active and involved
individual, not a passive or reclusive one. She would not
only read books and magazines, and occasionally try her
hand at poetry or fiction, but also participate in a varie-
ty of indoor, outdoor and community activities.
And authors wonder why they're sent to book-and-author
luncheons in Cleveland and Minneapolis! The explanation is sim-
ple: That's where the readers are— not merely the readers of Judith
Krantz and Belva Plain, but also the readers of Laurie Col win and
Gloria Naylor. The perceived image of the "literary" American
reader as a bearded male academic in a tweed jacket with leather
patches bears only scant connection to reality; the reader who really
supports serious American literature, such of it as there still may
be, is far more likely to be an educated woman of a certain age
who belongs to a neighborhood book club, buys her clothes through
the mail from Talbots, contributes to Greenpeace, and does volun-
teer work at the House of Ruth.
Though Zill and Winglee do not say so, there is a reason to
believe that this has been so for a couple of generations. But now
literature is beginning to catch up with its readers. "It can be ar-
gued," Zill and Winglee claim, "that the kinds of works being pub-
lished by literary presses in the United States today are very much
a reflection of the interests and concerns of this typical reader,"
and they are absolutely right.
If a single generalization can be made about contemporary
American literature, apart from its roots in the creative-writing
schools, it would be that it is the province of middle- and upper-
middle-class women. If in the past they were the readers, now they
are the writers as well. The point has been made before, but it
x Who Reads Literature?
is worth making again: While male writers of serious literature
under the age of 40 are notable in the United States largely for
their absence or their lack of consequence, female writers of their
generation are notable both for their numbers and for the quality
of their work. Unquestionably, Zill and Winglee are right:
The women's movement may be stimulating female in-
volvement with literature. Whenever norms and values
are in flux, literature has a special role to play. Litera-
ture can explore new patterns of behavior, provide charac-
ters that serve as role models, and give voice to both the
exhilaration and the frustrations that pioneers experience.
The drive for women's rights has helped to draw atten-
tion to outstanding women writers of the past and pres-
ent, and to open more opportunities for women in the
publishing and promotion of literature.
How large those opportunities will be is questionable at best:
The world of American literature is small by any standard, and
over the long haul is most unlikely to grow larger, in real if not
numerical terms. But it's a woman's world now, and if we are to
have in the next generation a literature of any consequence, it will
be because women make it so.
Jonathan Yardley
This article originally appeared in The Washington Post on August 28, 1989.
It is reprinted with the permission of the author and The Washington Post.
Introduction
Because the art of literature is inextricably linked to a country's
language and history, it has traditionally played a central role in
the culture of most nations. It is difficult to think of England with-
out thinking of Shakespeare and Dickens, or Russia without Push-
kin, Tolstoy, Chekhov, and Dostoyevsky. So it has been in the United
States, at least in the past. The American cultural heritage includes
characters, scenes, and phrases from the works of such authors
as Washington Irving, Edgar Allan Poe, Henry David Thoreau,
Herman Melville, Walt Whitman, Emily Dickinson, Mark Twain,
Ernest Hemingway, and Eugene O'Neill, among others.
Today, however, there is a widespread sense that the reading
of literature does not occupy a prominent place in the lives of most
Americans. Many observers feel that we are no longer "a nation
of readers," but a nation of watchers: watchers of movies, televi-
sion, videocassettes, and computer displays. Literary critic and
newspaper columnist Jonathan Yirdley complains about the "in-
creasing irrelevance of writing," and laments the fact that contem-
2 Who Reads literature?
porary American poets, from the laureate on down, are all but un-
known to the American people.1 University of Chicago professor
Allan Bloom, author of The Closing of the American Mind, as-
serts that "our students have lost the practice of and taste for read-
ing. They have not learned how to read, nor do they have the
expectation of delight or improvement from reading."2 Universi-
ty of Virginia professor E. D. Hirsch sounds a similar theme in
his book Cultural Literal contending that writers and speakers
can no longer take it for granted that young readers and listeners
will be familiar with works, characters, and authors that used to
be known by all educated people.3 There is even a new term,
"aliteracy," that has been coined to describe the phenomenon of
people who know how to read but choose not to do so.4
There is empirical evidence that seems to support these con-
tentions. For example, one national study found that U.S. adults
spend four times as much leisure time watching television or listen-
ing to the radio as they do reading books, magazines, or
newspapers.5 There are also survey results showing the American
public to be ignorant about basic literary matters. A 1984 Univer-
sity of Maryland survey found that only one quarter of American
adults knew who George Orwell, the celebrated author of the novel
1984, was.6 In 1986, the Educational Testing Service conducted
a national assessment of the literary and historical knowledge of
high school juniors. The Service found that less than 30 percent
of them could identify Tennessee Williams as the author of A Street-
car Named Desire, and less than a quarter knew something about
the plot of A Catcher in the Rye.1 Indeed, it seems possible that
students in the Soviet Union, who are known to be avid readers
of American literature, would do better at recognizing the works
of these and other modern American writers than students in the
United States.
Other evidence indicates, however, that for both reading in
general and literature reading in particular the situation may not
be quite so bleak. To begin with, there are a lot of books sold in
the United States: more than two billion each year during the
mid-1980s, or about nine books for every person over five years
of age. About 500 million of these are relatively inexpensive, "mass
market" paperbacks.8 Many of the paperbacks contain works of
Introduction 3
fiction, even if most of the titles might not qualify as what literary
critics would call literature.
Furthermore, the Educational Testing Service's assessment
found that virtually all high school students received some train-
ing in the appreciation of literature and are made to read at least
a few classic works by English and American authors. As a con-
sequence, today's students are still likely to know something about
Shakespeare, Dickens, Hawthorne, and F. Scott Fitzgerald.9 By
contrast, relatively few receive formal training in the visual arts,
music, or dance, and usually have only the foggiest of notions about
Jackson Pollock, George Gershwin, Charlie Parker, or Martha
Graham.10
But do young people go on to read literature when they leave
school and are no longer required to do so? Of all the arts, litera-
ture should be the one with the widest following. Only a minority
of young people learn to read music or play an instrument, draw
or paint proficiently, or act or dance on stage. But everyone who
is educable is expected to learn to read and write.11
Assessing the State of Literature Reading
We can infer from bestseller lists that American adults read real
estate investment guides, personal computer manuals, diet cook-
books, and the like. Do they also read novels, poetry, and plays?
The number of works of fiction published in the U.S. each year—
about 5,100 new titles or new editions12— suggests that some peo-
ple still read novels and short stories. This inference is reinforced
by the large number of fiction books sold each year— some 400
million copies through general retail outlets alone.13 To be sure,
many of the fiction titles published and sold are works of genre
fiction— thrillers, romances, science fiction, and the like— most
of which would not be considered works of high literature.
Nevertheless, even among the genre titles there are works written
with considerable craft and imagination, and read with enthusiasm
by people who could be spending their time watching movies or
television. In addition to the large output of fiction, there are about
1,000 volumes of poetry and drama published each year, and nearly
2,000 books of literary criticism and literary commentary.14
4 Who Reads Literature?
Clearly, the writing and reading of literature are not yet defunct.
But we need more than publication and sales figures to form
an accurate picture of how much literature reading is going on in
the United States. A person can buy a book without ever getting
around to reading it, or read a book that has not been bought, but
borrowed from a friend or a library. And literature is published
in periodicals as well as books. Thus we need information about
the reading habits of representative samples of American citizens,
including specifics about the kinds of books they read. Several large
surveys on participation in the arts or on book reading were car-
ried out in the United States during the 1980s for exactly this
purpose.
One of the most notable of these is the Survey of Public Par-
ticipation in the Arts (SPPA), which was a nationwide survey,
designed and sponsored by the National Endowment for the Arts
and conducted in 1982 and 1985 by the U.S. Bureau of the Cen-
sus. Issues not covered in the SPPA are examined in data drawn
from book industry publication and sales statistics, and from two
other national surveys: the Arts-Related Trend Study (ARTS) car-
ried out in 1983-1984 by the Survey Research Center at the Univer-
sity of Maryland, and the Consumer Research Study on Reading
and Book Purchasing done in 1983 for the Book Industry Study
Group (BISG).15
There are, of course, problems in using survey data to study
literature reading. Some of these problems are common to all sur-
veys that ask people to report on their own behavior, while others
are unique to studies of reading. British sociologist Peter Mann
points out that there are problems associated with research into
reading "which arise from the difficulty in determining what is
meant by 'reading' and what constitutes a 'book'. "16 Research on
the reading of literature is even more problematic because of dis-
agreements among experts on what should be included under the
rubric of "literature" and the difficulty of framing general and easily
understood questions about such reading.
Here, the reading of literature or what we will simply call
"reading" means the reading of novels, short stories, poetry, and
plays. As usually defined, literature also subsumes such high-quality,
non-fiction writing as essays, literary criticism, literary commen-
Introduction 5
tary, "belles lettres," biographies, and the so-called "non-fiction
novel." However, these forms of literature were not explicitly co-
vered in the surveys discussed here. In addition, distinctions be-
tween art and entertainment, based on the quality or seriousness
of the written work, are very important. Unfortunately, most of
these surveys do not include information that permits one to say
something about the quality of the books and magazine pieces that
American adults are reading. This kind of information was col-
lected in two of the other surveys described, but, even with these
data, drawing the line between literature and mere amusement is
no simple matter.
Because of the great expense and practical difficulties involved
in trying to observe directly the reading habits of large numbers
of Americans, the survey results presented here rely on people's
reports about the kinds of works they have read or not read within
broad intervals of time (the last 12 months or the last 6 months).
These reports are subject to both systematic bias and random er-
ror. To the extent that reading literature is perceived as something
that one "ought" to be doing, people will tend to say they have
read a novel or short story when, in fact, they have not. They may
also "telescope" events that happened in the past, such as reading
a book more than a year ago, and remember them as having oc-
curred within the reference period in question. On the other hand,
people tend to forget about things they did more than a few weeks
ago, especially if the event was not very important to them, and
this could result in underreporting. Accurate reporting also depends
on the respondent's understanding of what is meant by terms such
as "novel" and "short story," which may pose problems for less
educated individuals.
Two of the surveys described here attempted to get a sense
of the seriousness of some of these problems by asking respon-
dents follow-up questions. The answers to these questions provide
both further information about the works the survey respondents
have read, and a basis for adjusting estimates of the size of the
literature audience.
The survey situation does, however, have certain advantageous
aspects that are rarely encountered in everyday life: the respon-
dent is offered anonymity; honest reporting is explicitly encouraged;
6 Who Reads I ileralure?
and there is no overt praise or criticism for saying that one has
or has not done something. Moreover, a survey that uses scientif-
ic sampling procedures and achieves a high response rate provides
a picture of a real cross-section of the population, not just of a
limited and self-selected subset of people.
Even when there is an overall bias in survey reporting on an
activity, surveys can still provide an accurate reading of the com-
parative commonness of different forms of the activity, or of the
relative frequency of the activity among different groups, or of
changes in the frequency of the activity over time. It is known,
for example, that people tend to overreport voting in local or na-
tional elections: there are more people who say they voted than
the total number of ballots cast. Yet surveys of voting behavior still
give a good sense of the relative voting rates of different age, sex,
educational, ethnic, and residential groups, and show how these
patterns have changed over the last several decades.
In any event, self-report surveys, with all their limitations, are
the best source of information on literature reading that we have.
They will remain so until government or private groups invest in
studies that use direct observations of reading or ask for self-reports
that cover shorter time intervals and include more people in the
sample, and that elicit more extensive follow-up information on
the specific titles read.
Chapter 1
Readers of Fiction,
Poetry, and Drama:
How Many and
Who Are They?
The 1982 and 1985 rounds of the Survey of Public Participation
in the Arts (SPPA) took national probability samples of adults aged
18 and over living in households in the United States. These arts
surveys were done as supplements to larger survey programs in-
volving panels of respondents who were interviewed every six
months over a three-year period. In 1982, the SPPA interviewed
17,254 people, or 89 percent of the target sample. The sample was
about 20 percent smaller in 1985, when 13,675 people were inter-
viewed, or 85 percent of the target sample. Three quarters of the
interviews were done in person and the remainder by telephone.17
The SPPA interviews focused on attendance at arts exhibitions
and performances, including art museum shows, classical music
concerts, opera, jazz, plays, and musicals, and on other forms of
arts participation, including reading literature. In addition to a core
set of questions that were asked of all respondents, subsamples were
asked about training in the arts, mass media usage, and other forms
of leisure activity. A basic question on the reading of novels, short
8 Who Reads Literature?
stories, poetry, and plays was put to the entire sample in both sur-
veys, but questions on other forms of literature participation and
socialization were asked only of subsamples of about 4,200 to 5,500
respondents in 1982, and about 2,300 respondents in 1985.
In 1985, the SPPA found that 56 percent of a national sample
of adults aged 18 and over, representing 95.2 million people, report-
ed that they had read novels, short stories, poetry, or plays during
the last 12 months. The estimated number of readers was up by
nearly three million from the number in the 1982 SPPA. Howev-
er, this increase was the result of population growth only. The
proportion of adults who said they read literature was about the
same in both years.
The 1985 survey also asked whether respondents had read any
kind of book or magazine during the previous 12 months. Eighty-
six percent— representing some 146 million adults— said that they
had. If we divide the number of people who reported reading liter-
ature by the number who reported reading any kind of book or
magazine, we have an estimate of literature readers as a fraction
of all readers. In 1985, 65 percent of all adult readers in the U.S.
read some fiction, poetry, or drama in the course of a year.18 This
was slightly lower than the 67 percent found in the 1982 survey,
but the difference was within the margin of sampling error.
The SPPA collected additional information on public partici-
pation in one particular form of literature— poetry. In the 1985 sur-
vey, 19 percent of the respondents— representing 32 million
adults— reported that they had read or listened to a reading of po-
etry during the previous 12 months. For the 1982 survey, these
figures were 20 percent and 30 million, respectively. Again, the
differences were not statistically significant.
In addition to questions about reading literature, the SPPA asked
respondents if they had worked on "any creative writings, such
as stories, poems, plays, and the like" during the last 12 months.
There was no requirement that the writing had been published,
and the results of an independent follow-up study indicate that most
of it probably was not. In the 1985 survey, 6 percent of the
respondents — representing 10.6 million adults — said that they had
tried to do some creative writing. This was about the same as the
1982 results, when 7 percent— representing 10.7 million adults—
Readers of Fiction, Poetry, and Drama 9
answered the question affirmatively. The apparent decline in the
proportion of writers was not statistically significant.
Comparing SPPA Results with Other Surveys and Sales Figures
The levels of reading reported in the SPPA are in at least ap-
proximate agreement with the results of other nationwide surveys.
For example, the Consumer Research Study on Reading and Book
Purchasing conducted in 1983 for the Book Industry Study Group
(BISG) found that 39 percent of all adult respondents had read a
book of fiction in the last six months and half had read a book
of some sort; and 92 percent had read magazines, periodicals, or
newspapers over the same period.19 Similar results have been ob-
tained in other countries. In Britain, for instance, a number of na-
tional studies done in the late 1970s and early 1980s found that,
as in the U.S., roughly half the adult population reported reading
books of one sort or another. In a 1981 Euromonitor survey of about
2,000 people aged 16 and over, 45 percent said they were reading
a book (any book) at the time of the survey and about 30 percent
said they were reading a work of fiction.20
Despite the general agreement among readership studies, their
estimates are typically met with incredulity by those involved with
the writing, publishing, or support of contemporary literature. What
literary people point out is that it is not uncommon for a work of
serious fiction to sell fewer than 5,000 copies nowadays. Likewise,
the circulation of most poetry magazines is counted in the low thou-
sands or even hundreds. In 1987, the Los Angeles Times Book Re-
view announced that it would no longer be reviewing new volumes
of poetry because there was so little reader interest in them. Even
the most widely read magazines that publish first-rate fiction and
poetry— magazines like The New Yorker and The Atlantic— have
circulations in only the 400,000-600,000 range.21
John P. Dessauer, a leading expert on book industry sales
trends, has estimated that a total of 3.2 million copies of contem-
porary literary fiction and poetry books were sold through gener-
al retailers in 1985, representing just 0.3 percent of all books sold
through these outlets. Sales of classic works of literature made up
another 9.1 million units, or 0.9 percent of books sold. Thus, con-
temporary and classic literature together constituted little more than
10 Who Reads Literature?
one percent of bookstore sales.22 If there are so many readers of
literature out there, why do the literary books and magazines not
sell better?
Of course, people get reading material from friends and rela-
tives, from public libraries, in doctors' and dentists' offices, and
from their own stock of books acquired over the years. When fic-
tion readers surveyed in the BISG study were asked where they
had obtained the last book they had read, less than half— 45
percent— said they had purchased it themselves. More than a quarter
said they borrowed the book from a friend or relative or traded
it for another book. Another fifth had borrowed the book from
a library, and 5 percent had received it as a gift.23 However, even
doubling or tripling the estimated number of literature books sold
to account for books borrowed and exchanged would not bring the
total close to the 95 million readers that the SPPA found.
What does bring the survey and book sales figures into line
with one another is incorporating the large numbers of copies of
romances, thrillers, science fiction novels, and other works of popu-
lar or genre fiction that are sold each year. John Dessauer esti-
mates that total sales of "popular fiction" books through general
retail outlets amounted to more than 322 million copies in 1985.
And that does not include nearly 124 million in "bestseller" sales.
(Assuming that about two-thirds of the bestsellers were fiction would
bring the total number of popular fiction books sold through general
retail outlets to about 400 million.)24
Thus, what most of the survey respondents seem to be talking
about when they report that they have read novels or short stories
are works of relatively light, genre fiction. Inasmuch as many of
these works would not qualify as literature in the eyes of most liter-
ary critics, the implication is that the adult audience for serious
contemporary literature is probably a good deal smaller than the
56 percent found in the SPPA. These impressions are strengthened
by survey information on the specific titles or the kinds of works
to which people are referring when they report that they have read
fiction, poetry, or drama. Information on works read was not col-
lected in the Survey of Public Participation in the Arts, but rele-
vant data are available from the Arts-Related Trend Study (ARTS)
conducted by the University of Maryland and the Consumer Re-
Readers of Fiction, Poetry, and Drama 1 1
search Study on Reading and Book Purchasing done for the Book
Industry Study Group. These data are examined later.
What Kinds of People Read Literature?
Demographic Characteristics
People who report reading fiction, poetry, and drama are a
diverse group. They are found in every segment of the U.S. popu-
lation, except those subgroups who do not read at all. However,
some segments of the population are overrepresented among readers
(and writers) of literature:
• those who have at least some college education (who make
up 49 percent of literature readers, as opposed to 36 per-
cent of the general population);
• those with incomes of $25,000 and over (who comprise
48 percent of literature readers, but 40 percent of the general
population);
• females (59 percent of literature readers, but 53 percent
of the general population);
• the middle-aged (40 percent of literature readers, but 36
percent of the general adult population);
• whites (85 percent of literature readers, but 81 percent of
the general population).
Conversely, groups that are underrepresented among litera-
ture readers include the following:
• those with less than a high-school education (who com-
prise 14 percent of literature readers, but 25 percent of the
general population);
• those with incomes under $10,000 (16 percent of literature
readers, 21 percent of the general population);
• males (41 percent of literature readers, 47 percent of the
adult population);
• those aged 50 and older (32 percent of literature readers,
35 percent of the general adult population);
1 2 Who Reads Li terature?
•
Blacks and Hispanics (13 percent of literature readers, 17
percent of the general population).
As one goes from the overall population, to those who read
books and magazines, to those who read literature, to those who
read or listen to poetry, and to those who try to produce creative
writing, the groups become progressively more college-educated,
more female, and more middle-income. (Table 1.) Thus, of the self-
described writers in the 1985 SPPA, 69 percent were college-
educated, 63 percent were female, and 51 percent had incomes of
$25,000 or more. (Given that most of the writing reported in the
SPPA was probably unpublished, and that even when published,
writing is usually not handsomely rewarded, we can be confident
that the income of these creative writers came primarily from
sources other than their writings.)
The relationships between literary participation and personal
characteristics, such as education, income, age, sex, and race, as
well as the reasons behind the observed relationships, are exam-
ined in greater detail later.
Geographic Distribution
The writing, publishing, and reading of literature are often
thought of as Northeastern, big-city enterprises. But the arts sur-
vey data show that these readers and writers are spread through-
out the four major regions of the country, pretty much in line with
the distribution of the total adult population. (Table 1.) If any re-
gion was overrepresented, it was the West. In 1985, for example,
the West contained 19 percent of the overall adult population, but
had 22 percent of the readers and 33 percent of the writers of liter-
ature. The Midwest, with 25 percent of the adult population, had
26 percent of readers and 30 percent of writers. The South tended
to be underrepresented in this regard. But, being the largest re-
gion in terms of overall population, the South contained nearly a
third of all readers and almost a quarter of all writers.
The majority of readers and writers do live in the large
metropolitan areas of the country. But, like the rest of the more
educated and affluent population, most of them live in the suburbs,
not the central cities. In 1985, the suburbs held 41 percent of the
Readers of Fiction, Poetry, and Drama 13
TABLE I. Size and Composition of U.S. Population of
Writers of Literature, by Age, Gender, Ethnic Group,
Region, and Metropolitan Residence, U.S. Adults Aged
Adult Readers and
Education, Income,
18 and Over, 1985.
Total
Creative
Adult
All
Literature
Poetry
Literature
Population
Readers
Readers
Readers
Writers
No. in population
(in millions)
170.6
146.0
95.2
31.8
10.6
% of Adult Pop.
100%
86%
56%
19%
6%
% of All Readers
—
100%
65%
22%
7%
TOTAL
100%
100%
100%
100%
100%
Percent Distribution
AGE
Young (18-29)
28%
29%
29%
30%
39%
Middle (30-49)
36%
38%
40%
38%
44%
Older (50 +)
35%
33%
32%
32%
17%
GENDER
Female
53%
54%
59%
60%
63%
Male
47%
46%
41%
40%
37%
ETHNIC GROUP
White
81%
85%
85%
86%
87%
Black
11%
8%
8%
8%
8%
Hispanic
7%
5%
5%
5%
4%
Asian, Other
2%
2%
2%
2%
1%
EDUCATION
Some College
36%
42%
49%
56%
69%
High School Grad
39%
38%
37%
30%
23%
Less than HS
25%
20%
14%
15%
8%
INCOME
$25K & over
40%
45%
48%
47%
51%
$I0-25K
39%
38%
36%
39%
39%
Under $ 1 OK
21%
17%
16%
15%
10%
REGION
Northeast
21%
19%
21%
17%
13%
Midwest
25%
27%
26%
32%
30%
South
34%
32%
31%
33%
24%
West
19%
23%
22%
19%
33%
RESIDENCE
Central City
27%
25%
27%
26%
32%
Suburbs
41%
45%
45%
45%
53%
Non-Metro
32%
30%
28%
29%
15%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
1 4 Who Reads Literature?
general adult population, but 45 percent of the readers and 53 per-
cent of the writers. The comparable figures for the central cities,
with 27 percent of the population, are 27 and 32 percent, respec-
tively. People living outside of metropolitan areas were under-
represented: these areas contained 32 percent of the adult population
in 1985, but only 28 percent of the literature readers and just 15
percent of the writers.
Leisure Activity Profile
In the report on the 1983 Book Industry Study Group (BISG)
survey of book reading habits, the following note was made:
Book readers are often portrayed in literature, films, or
on stage as solitary, somewhat aloof, self-absorbed per-
sonalities whose devotion to their books seems to take
the place of interaction with the rest of the world. This
study, however, proves the stereotype to be nothing more
than a myth. Far from being introverted or social out-
casts, book readers emerge as well-rounded individuals
active in a wide range of social and cultural activities.25
The BISG study found that book readers were more active than
non-book readers in many areas, including that of socializing with
others.
The SPPA obtained a very similar result. In addition to infor-
mation about literature reading and arts attendance, the SPPA col-
lected data on participation in a variety of other leisure activities
during the 12 months prior to the survey. When the reports on
recreational activities were cross-tabulated with the measures of
literary participation, it was found that people who had read fic-
tion, poetry, and drama in the last year were more active in virtu-
ally all areas than people who had done reading, but not of literature.
The latter group was more active, in turn, than those who had not
read any books or magazines at all. (Table 2.)
Literature readers were not only more active in areas where
one might expect them to be (e.g. , visiting arts fairs, historic sites,
and museums; doing gardening or gourmet cooking; or taking part
in arts and crafts activities), they were also more active in going
to less refined amusement events: playing games and sports; tak-
Readers of Fiction, Poetry, and Drama 15
TABLE 2. Leisure Activity Profile of Total Adult Population, Literature Read-
ers, Non-Literature Readers, and Non-Readers, U.S. Adults Aged 18 and Over,
1982.
READERSHIP GROUPS
Total
Readers,
Adult
Literature
But Not of
Non-
Population
Readers
Literature
Readers
Proportion of Group
That Has Done
LEISURE ACTIVITIES
Activity in Last 12 Months
Amusements
Play card, board games
65%
77%
62%
27%
Attend movies
63%
75%
59%
25%
Visit amusement park
49%
57%
49%
19%
Attend sports events
48%
59%
43%
17%
Exercise, Sports
Jog, exercise
52%
65%
43%
18%
Play sports
39%
48%
36%
14%
Camping, hiking
37%
43%
34%
14%
Home-Based Activities
Repair home, car
60%
66%
60%
28%
Gardening
61%
69%
53%
34%
Gourmet cooking
29%
38%
22%
8%
Collect stamps, coins
15%
20%
10%
3%
Charitable Activities
Volunteer, charity work
28%
36%
21%
9%
Cultural Attendance
Visit art/crafts fairs
39%
54%
28%
10%
Visit historic sites
37%
50%
28%
8%
Go to zoo
32%
41%
25%
11%
Visit science, natural
history museums
23%
32%
15%
4%
Art & Crafts Activities
Weaving, needlework
33%
42%
29%
18%
Pottery, ceramics
13%
17%
9%
3%
Photography, video
10%
14%
6%
2%
Painting, drawing,
sculpture, printmaking
10%
14%
6%
2%
Backstage theatre help
3%
4%
1%
0%
READERSHIP GROUP SIZE
% of Adult Population 100%
57%
26%
5%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
1 6 Who Reads Literature?
ing part in outdoor activities; doing home and car repairs; and con-
tributing their time to charity. For example, three quarters of the
literature readers had gone to the movies in the last year, whereas
less than 60 percent of the non-literature readers and only a quar-
ter of the non-readers had done so. Two thirds of the literature read-
ers had done jogging or other similar exercise, whereas less than
half of the non-literature readers, and less than a fifth of the non-
readers, had participated in some form of exercise program. More
than a third of the literature readers had done volunteer or charity
work, compared with a fifth of the non-literature readers and a
tenth of the non-readers.
The higher activity levels of the literature readers were partly
a function of their being better educated, more affluent, and youn-
ger, on the average, than their counterparts. There may also have
been an element of shared reporting bias in the associations, in
the sense that respondents who were more likely to remember and
report one kind of activity were more apt to remember and report
other kinds as well. Nonetheless, there does seem to be a genuine
link between literature reading and other cultural and recreational
activities.
It is not that reading literature caused the other activities, or
vice versa. Rather, individuals seem to differ in their overall curi-
osity and activity levels, and those who have the interests and energy
to do one kind of cultural or recreational activity are more likely
to do others also. In some cases, there is a common thread linking
literature reading with other activities, as when an individual has
an interest in the Civil War, and reads historical novels about that
period, visits Civil War battle sites, and goes to military muse-
ums. Even lacking a common interest, however, the operative prin-
ciple seems to be the more, the more, rather than one activity versus
the other.26
Thumbnail Sketch of the Literature Reader
In sum, if we had to put together a picture of a typical reader
of literature in the United States today, the survey data indicate that
the person would be a middle-aged white female living in the
suburbs of a Western or Midwestern city. She would have a col-
lege education, and a middle- to upper-middle class income that
Readers of Fiction, Poetry, and Drama 17
was not derived from her literary activities. She would be an ac-
tive and involved individual, not a passive or reclusive one. She
would not only read books and magazines, and occasionally try
her hand at poetry or fiction, but also participate in a variety of
indoor, outdoor, and community activities.
Obviously, there are many readers and creative writers who
do not conform to this stereotype. Indeed, one of the heartening
aspects of the contemporary literary scene is its ethnic and cultur-
al diversity. Nonetheless, it can be argued that the kinds of works
that are being published by literary presses in the U.S. today are
very much a reflection of the interests and concerns of this typical
reader.
Is Literature Reading Growing or Diminishing?
There are a number of reasons for believing that the audience for
literature should be growing. As the U.S. population gradually
changes, older cohorts are being replaced by those whose parents
had more education and were more apt to have encouraged their
children to read. The younger cohorts have also had more years
of schooling and are more likely to have been exposed to creative
writing courses. As shown later, all of these factors are positively
associated with literature reading as an adult. Thus, while we could
expect that there should be more literature reading occurring in
the future, it is not clear that this growth will really take place.
Taken together, the 1982 and 1985 rounds of the SPPA indicate
that the proportion of literature readers is holding steady, while
the number of readers is growing with the overall population. How-
ever, because the two rounds of the survey are separated by just
three years, we can glean only a limited picture of the longer-range
changes that may be taking place. It should be possible to get a
clearer view of long-term trends by viewing the SPPA findings in
conjunction with the results of other surveys and book sales data
from the publishing industry.
Book Reading: Past Growth, Recent Decline
It does seem to be the case that a greater proportion of the
public reads books now than did so several decades ago. Data from
18 Who Reads Literature?
Gallup polls conducted in 1955 and 1984 show a 50 percent in-
crease over that period in the proportion of respondents who report-
ed that they had read a book (other than the Bible) "yesterday."
The proportion grew from 14 to 21 percent, with much of the in-
crease attributable to the expansion in the portion of the popula-
tion that was college educated.27 Data from the SPPA also show
that middle-aged adults do more general reading and more litera-
ture reading than older adults. (Table 3.) As demonstrated later,
these differences seem to represent an historical increase in read-
ing over successive generations rather than a decline in reading
with age. But is the increase continuing? Although reading in gener-
al still seems to be growing, there is evidence to indicate that book
and literature reading are not. Indeed, among young adults these
forms of reading may actually be on the decline.
Evidence of a recent decline in book reading comes from two
national surveys sponsored by the Book Industry Study Group
(BISG) . The surveys were conducted in 1978 and 1983. Whereas
overall reading (including newspapers and magazines) was stable
over that period, there was a 5 percentage point reduction in the
proportion of adults who had read books in the previous six months.
More ominously, the proportion of book readers among young
adults (ages 16-20) dropped by 13 points, from 75 to 62 percent.28
Trends in Book Sales
Indications that literature reading represents a diminishing share
of all book reading can be found in sales figures from the publish-
ing industry. Whereas the total number of books sold each year
in the U.S. grew from 1.5 billion copies in the mid-1970s to more
than 2 billion in the mid-1980s, unit sales of mass market paper-
backs remained fairly stationary, at about 500 million copies an-
nually29 Mass-market paperbound books are, of course, the form
in which much popular fiction is published or reprinted. Although
sales of higher-priced "trade" paperbounds* have grown, trade books
in general are capturing a decreasing share of the U.S. book mar-
ket. Technical, scientific, professional, and reference works are
*As used here, the term "trade books" includes fiction and general-interest non-
fiction in hard cover and higher-priced paperbound editions, juvenile books, and
mass market paperbacks.
Readers of Fiction, Poetry, and Drama 19
capturing an increasing share of the market. While total annual
book sales in the U.S. grew from $2.3 billion in 1968 to a project-
ed $12.8 billion in 1988, the trade book segment of the market
declined from 30 to 23 percent over the same period.30
Declining Reading by Young Adults
Figures from the two SPPA studies indicated constancy, rath-
er than decreases, in the overall proportion of literature readers
in the population. (Table 3.) Their data suggested declines in po-
etry and writing, but the observed changes may be due to sam-
pling fluctuations. Among those under 30, however, there were
statistically significant changes between the two surveys: literature
reading dropped from 61 to 57 percent; poetry reading fell from
24 to 20 percent; and overall reading declined from 89 to 87 per-
cent. Although these differences may seem small, they would be-
come considerable if the same rates of decrease were to continue
over a longer period.
The data from the SPPA and BISG findings reported above
are not the only signs of less frequent reading among young adults.
Data from an annual, school-based survey of high school seniors
called Monitoring the Future shows a gradual diminution in the
proportion who report reading books, magazines, or newspapers
"almost every day," from 62 percent in 1977 to 46 percent in
1988.31 Thus, evidence from three different survey programs
points to the conclusion that a decline in reading is occurring among
successive cohorts of young adults in the United States.
A Fluid Situation
Why is literature reading remaining stagnant or even declin-
ing, when various demographic factors indicate that it should be
increasing? Reasons for the lack of growth are examined at the con-
clusion of this monograph. We note here, though, that the situa-
tion is a fluid one, especially as far as sales of literature are
concerned. With so many potential readers in the population, and
such a small fraction of them needed to make a bestseller, there
could be short-term increases in literature sales even while a long-
term decline in literature reading was in progress. Book sales also
depend on economic conditions, the popularity of the current crop
20 Who Reads Literature?
TABLE 3. Change in Proportion of Adult Population and Population Subgroups
That Have Read Literature, Read Books or Magazines, Read Poetry, and Done
Creative Writing in the Last 12 Months, U.S. Adults Aged 18 and Over, 1982
to 1985.
Literature Readers
All Readers
Differ-
Differ-
All Adults (18+)
1985
1982
ence
1985
1982
ence
No. in population
(in millions)
95.2
92.5
2.7
146.0
138.0
8.0
% of Adult Pop.
56.0%
56.4%
-0.4%
85.6%
84. 1 %
1 .4%
% of All Readers
65.2%
67.0%
-1.8%
100.0%
100.0%
—
Population Subgroups
AGE
Young (18-29)
56.8%
60.9%
-4. 1 %
87.0%
89.4%
-2.4%
Middle (30-49)
60.8%
59.7%
1.1%
88.6%
87.4%
1 .2%
Older (50 +)
50.3%
49.6%
0.7%
81.5%
75.9%
5.6%
GENDER
Female
63.0%
63.0%
0.0%
88.3%
85.6%
2.7%
Male
48. 1 %
49. 1 %
-1.0%
82.7%
81.8%
0.9%
ETHNIC GROUP
White
59.0%
59.8%
-0.8%
89.9%
86.4%
3.5%
Black
43.0%
42.3%
0.7%
66.3%
71.3%
-5.0%
Hispanic
41.5%
36.4%
5. 1 %
66.0%
72.2%
-6.2%
Asian, Other
51.9%
50.2%
1.7%
85.3%
80.2%
5. 1 %
EDUCATION
Some College
75.4%
77.7%
-2.3%
97.2%
96.6%
0.6%
High School Grad
53.4%
55.4%
-2.0%
85.9%
88.0%
-2. 1 %
Less than HS
32.6%
31.2%
1 .4%
68.4%
63.7%
4.7%
INCOME
$25K & over
66.5%
69. 1 %
-2.6%
92.3%
94.2%
-1.9%
SI0-25K
51.8%
55.0%
-3.2%
85.3%
85.4%
-0. 1 %
Under $ 1 OK
43.6%
43.2%
0.4%
72.3%
69.6%
2.7%
REGION
Northeast
57.0%
58.3%
-1.3%
86.4%
84. 1 %
2.3%
Midwest
56.7%
58.4%
-1.7%
90.3%
88.8%
1.5%
South
50.4%
49.0%
1.4%
80.6%
76.6%
4.0%
West
63.7%
63.9%
-0.2%
87.2%
89.0%
-1.8%
RESIDENCE
Central City
56.5%
56.5%
0.0%
85.5%
83.4%
2. 1 %
Suburbs
61.0%
60.2%
0.8%
91.2%
88.5%
2.7%
Non-Metro
48.9%
51.7%
-2.8%
78.4%
78.5%
-0. 1 %
(continued)
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions by N. Zill and M. Winglee from public use data files.
Readers of Fiction, Poetry, and Drama 21
TABLE 3. (Continued) Change in Proportion of Adult Population and Popula-
tion Subgroups That Have Read Literature, Read Books or Magazines, Read
Poetry, and Done Creative Writing in the Last 1 2 Months, U.S. Adults Aged
18 and Over, 1982 to 1985.
Poetry Readers
Creative Writers
Differ-
Differ-
All Adults (18+)
1985
1982
ence
1985
1982
ence
No. in population
(in millions)
31.8
32.5
-0.7
10.6
10.7
-0.1
% of Adult Pop.
18.6%
19.8%
-1.2%
6.2%
6.5%
-0.3%
% of All Readers
21.8%
23.6%
-1.8%
7.3%
7.8%
-0.5%
Population Subgroups
AGE
Young (18-29)
19.7%
24.0%
-4.3%
8.5%
10.5%
-2.0%
Middle (30-49)
20. 1 %
21.1%
-1.0%
7.6%
6.6%
1 .0%
Older (50 + )
17.4%
15.2%
2.2%
3.0%
3.1%
-0. 1 %
GENDER
Female
21.5%
23.0%
-1.5%
7.4%
8.1%
-0.7%
Male
16.2%
16.2%
0.0%
4.9%
4.7%
0.2%
ETHNIC GROUP
White
20.0%
20.5%
-0.5%
6.7%
6.6%
0. 1 %
Black
13.8%
15.1%
-1.3%
4.5%
5.7%
-1.2%
Hispanic
14.8%
16.9%
-2. 1 %
4.0%
7.0%
-3.0%
Asian, Other
16.0%
23.1%
-7. 1 %
2.4%
6. 1 %
-3.7%
EDUCATION
Some College
28. 1 %
3 1 .0%
-2.9%
1 1.5%
1 1.6%
-0. 1 %
High School Grad
14.6%
17.9%
-3.3%
3.8%
4.7%
-0.9%
Less than HS
12.1%
8.0%
4. 1 %
2.0%
2.6%
-0.6%
INCOME
$25K & over
22.6%
24. 1 %
-1.5%
8.0%
7.4%
0.6%
$I0-25K
19.6%
18.8%
0.8%
6.4%
6.0%
0.4%
Under $ 1 OK
14.1%
16.6%
-2.5%
3.2%
5.5%
-2.3%
REGION
Northeast
17.1%
19.5%
-2.4%
4.6%
6.5%
-1.9%
Midwest
21.0%
20.7%
0.3%
6.6%
5.5%
1.1%
South
18.5%
17.0%
1.5%
4.6%
5.6%
-1.0%
West
17.0%
23.3%
-6.3%
10.0%
9.4%
0.6%
RESIDENCE
Central City
18.5%
20.7%
-2.2%
7.5%
8.4%
-0.9%
Suburbs
21.2%
20.0%
1.2%
8.2%
6.6%
1.6%
Non-Metro
16.7%
18.9%
-2.2%
2.8%
4.9%
-2. 1 %
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions by N. Zill and M. Winglee from public use data files.
22 Who Reads Literature?
of authors and titles, and promotional and marketing factors. Partly
because of the positive demographic omens mentioned above, the
U.S. Department of Commerce is forecasting healthy growth in
the book publishing industry through the early 1990s.32
The prospects for literature readership depend on whether the
observed declines in reading among young adults continue, and
on the balance between the older portion of the population (where
literature reading seems to be growing) and the younger portion
(where it seems to be declining). The current middle-aged popu-
lation (who were products of the post-war "baby boom") is rela-
tively large, and the young adult population (who were products
of the "birth dearth" years) relatively small. Thus, although there
is cause for concern about the long-term future of literature, there
is reason for guarded optimism in the short run.
Chapter 2
What the Readers Are Reading
Two studies gathered information not only on whether people had
read fiction, poetry, or drama, but also on the specific kinds of
works they read. The Arts-Related Trend Study asked respondents
for specific examples of works they had read, and classified these
according to their literary quality, and how appropriate and con-
temporary they were. The other study, a survey done for the Book
Industry Study Group, did not ask for specific titles, but inquired
whether the respondent's reading included various forms and genres
of fiction, such as mysteries, romances, science fiction, etc. These
studies give a more detailed picture of the kinds of reading Ameri-
cans are doing, and they permit us to make a rough estimate of
the size of the audience for serious, as opposed to popular, literature.
Asking for Titles
The Arts-Related Trend Study (ARTS), a nationwide telephone
survey on arts knowledge and participation, was conducted by the
Survey Research Center at the University of Maryland in June 1983
24 Who Reads Literature?
and January 1984. 33 The sample interviewed for this study (1,077
adults) was considerably smaller than the samples surveyed in the
1982 and 1985 rounds of the SPPA, and its completion rate (70
percent) was lower. But the study collected illuminating follow-up
information on the kinds of arts-related activities reported by the
SPPA respondents, including the titles and authors of some of the
works of literature that each respondent had read during the previ-
ous 12 months. When categorized and tabulated, this sample of
works read begins to give us a picture of what people mean when
they report that they have read literature recently.
The proportion of respondents reporting that they had read one
or more works of fiction, poetry, or drama during the previous
12 months was similar to that found in the 1982 SPPA, although
about 4 percentage points lower. In addition to the combined ques-
tion about reading novels, short stories, poetry, or plays, the Univer-
sity of Maryland surveys asked separately about each of these
categories of literature.
Novel Reading
Forty percent of the respondents reported that they had read
one or more novels during the last 12 months. When asked to give
some examples of novels they had read, however, nearly a quarter
of the self-described readers could not come up with the name of
a specific book or author, or gave the name of a work that was
not a novel, but a biography, self-help book, or other non-fiction
title. Another 30 percent named only works of light, popular fic-
tion, such as a "blockbuster" by Judith Krantz or Sidney Shel-
don, a horror story by Stephen King, a romance by Victoria Holt,
a western by Louis L' Amour, a novelization of one of the "Star
Wars" films, etc. Ten percent of the novel readers named a classic
work, such as a novel by Dickens, Tolstoy, Henry James, Mark
Twain, or Hemingway. Seventeen percent reported reading a con-
temporary work of some literary merit, such as William Styron's
Sophie's Choice, Norman Mailer's Ancient Evenings , Alice Walk-
er's The Color Purple, or John Updike's Couples.
In terms of overall percentages, 30 percent of all U.S. adults
reported reading novels in the last 12 months and could give at
least one name that qualified as a title or author of an actual nov-
What the Readers are Reading 25
el. Only about 11 percent of all adults seemed to have read a work
of some literary distinction * however, and only 7 percent had read
a meritorious contemporary work. The latter figure is remarkably
close to a figure reported by Peter Mann, namely, that 6 percent
of British adults who were found to be reading "modern novels"
in the 1981 Euromonitor readership survey in Great Britain.34
Short Story Reading
Twenty-eight percent of the respondents to the ARTS survey
reported reading short stories during the previous twelve months.
However, when asked to recall the authors or titles of some of these
stories, or the name of the magazine or book in which the stories
appeared, many had difficulty. More than a quarter of the ostensi-
ble story readers could not provide any descriptive information
about the stories, or gave the titles of inappropriate works. Anoth-
er 10 percent gave responses that could not be classified. Nearly
45 percent more gave only the name of the magazine in which the
story appeared, and many of these magazines were ones which con-
tained non-fiction as well as fiction (e.g., Reader's Digest, Red-
book, Family Circle), or non-fiction feature stories only (Newsweek,
National Geographic). Thus, there seemed to be confusion in some
respondents' minds as to what the term "short story" signified.
Less than 20 percent of the story readers named authors, stories,
or anthologies of stories that could be classified as "serious" liter-
ature; only 5 percent named contemporary writers or stories of
literary merit.
In terms of overall percentages, 20 percent of all U.S. adults
reported reading short stories and could give some descriptive in-
formation about the stories. But only 5 percent of all adults had
read stories that could be ascertained to be of literary quality, and
less than two percent had read contemporary short stories of liter-
ary value.
Judgments about the literary merit of various works are arguable, of course.
The categorizations reported here are those made by the staff of the Maryland
Survey Research Center, presumably after some consultation with faculty experts
on literature. For the most part, these categorizations seem reasonable, although
a perusal of the actual responses, which are listed in an appendix to the survey
report, reveals some anomalous classifications and a few coding errors.
26 Who Reads Literature?
Poetry Reading
Fifteen percent of the adults surveyed in the Arts-Related Trends
Study reported reading poetry during the past 12 months. This was
5 percentage points lower than the proportion reported in the 1982
SPPA * When asked to provide the names of poets or poems read,
or the title of the magazine or book in which the poems were found,
nearly 70 percent of the poetry readers were able to provide some
corroborative detail. But almost a quarter gave only the name of
a mass-circulation magazine such as Parade or Reader's Digest,
or named examples of less serious forms of verse, such as "Gross
Limericks," popular song lyrics, or poems written for children.
On the other hand, close to 40 percent of the poetry readers named
poets, poems, and/or poetry anthologies of literary distinction, in-
cluding works by T.S. Eliot, Robert Frost, Emily Dickinson, Carl
Sandburg, Ezra Pound, Edgar Allan Poe, Robert W. Service, Henry
Wadsworth Longfellow, and William Carlos Williams. Very few
of the names or poems mentioned were those of serious living poets,
however.
As a proportion of the total population, 10 percent of U.S. adults
reported reading poetry and could provide some information on
what or where poems were read. Six percent had read poems of
clear literary merit, mostly modern or traditional classics. One per-
cent or less had read serious contemporary poetry.
Play Reading
Although only 5 percent of the adults surveyed in the ARTS
reported reading a play during the previous 12 months, 91 percent
of them could name a specific play or dramatist, or both. More-
over, 80 percent of the authors and works mentioned seemed to
have literary merit, although less than 10 percent of them were works
of living playwrights. Examples of names or plays mentioned in-
clude those of Shakespeare, Shaw, Tennessee Williams, Brecht,
Lillian Hellman, Tom Stoppard, and Ntozake Shange. In terms of
the total adult population, 5 percent reported reading plays and could
*The difference suggests that follow-up questions may have had a suppressing
effect on the reporting of literary participation. Because this kind of effect is com-
mon in survey research, it is good practice to ask all screening questions before
asking any follow-up questions. This was not done in the ARTS survey.
What the Readers are Reading 27
give the name of a specific play or playwright read. Four percent
had read drama of literary merit, but less than one percent had
read serious contemporary dramas.
Table 4 summarizes the ARTS findings on novel, short story,
poetry, and play reading. Unfortunately, the published results do
not indicate how much overlap there was across these types of read-
ing, so estimates of the total size of the audience for works of literary
merit can only be approximate. Depending on the degree of over-
lap assumed, the total proportion of people reading works of mer-
it could range from a little more than 10 percent up to 25 percent
or more, whereas the proportion reading contemporary works of
merit could range from 7 to about 10 percent.
Creative Writing
The ARTS survey also asked more detailed questions than the
SPPA about creative writing activity. The initial question was, "In
the last 12 months, have you taken any lessons in creative writing
or done any creative writing for your own pleasure"? If the respon-
dents indicated that they had, they were asked what types of work
they had tried to write (stories, novels, poetry, or plays) and whether
they had written anything that had been published. All ARTS
respondents were also asked if they felt they were able to do crea-
tive writing.
Nine percent of the arts respondents said they had written or
taken writing lessons in the last 12 months. This was higher than
the 7 percent who reported doing creative writing in the 1982 SPPA,
but the comparable SPPA question did not include writing lessons.
Poetry writing was the most common form mentioned; it was at-
tempted by 6 percent of adults (or 62 percent of those who did
some writing). Work on stories or novels was reported by 4 per-
cent of adults (or 38 percent of the writers). Play writing, which
was reported by one percent of the adults (or 9 percent of the
writers), was least common.
Only about a quarter of the writers, or 2 percent of all respon-
dents, said they had had something published. This included pub-
lication in relatively informal outlets such as school magazines,
organizational newsletters, etc. More than a fifth of all the ARTS
28 Who Reads Literature?
TABLE 4. Proportions of U.S. Adult Population That Report Reading Various
Forms of Literature in Last 12 Months, and Proportions Reading Works of
Literary Merit, U.S. Adults Aged 18 and Over, 1983-84.
Have Read
Works in This
Can Provide
Mention
Mention Con-
Form in
Information
Work or
temporary
Last 12
About
Author of
Work of
Literary Form
Months
Works Read
30%
Literary Merit
1 1%
Merit
Novels
40%
7%
Short Stories
28%
20%
5%
1%
Poetry
15%
10%
6%
1%
Plays
5%
5%
4%
>l%
SOURCE: Developed from data in: Robinson, John R. et at., Americans' Par-
ticipation In The Arts: A 1983-84 Arts-Related Trend Study. Final
Report, College Park, MD: University of Maryland Survey Research
Center, 1986.
What the Readers are Reading 29
respondents— 22 percent— felt that they had the ability to do crea-
tive writing.
Varieties of Fiction
Information about the kinds of works that are read by literature
readers was also collected in the 1983 Consumer Research Study
on Reading and Book Purchasing conducted for the Book Indus-
try Study Group.35 Instead of asking for specific titles and
authors, the BISG survey inquired about categories of fiction read,
covering various genres of novels, as well as short stories, poetry,
and drama under the fiction rubric. There was no attempt to evaluate
the literary quality of the works. The survey used a six-month
reporting period, as opposed to the 12-month period used in the
SPPA or ARTS questionnaires. The BISG questions about the types
of fiction read were only asked of those who reported reading at
least one fiction book during the reference period.
Genre Fiction
The BISG survey found that the novel was the most widely
read form of fiction. However, much of the novel reading was spread
across a variety of popular genres that are not usually thought of
as "literary," though they occasionally produce individual works
or authors of enduring quality. Each genre accounted for between
10 and 40 percent of all fiction readers, or about 4 to 15 percent
of all adults. As indicated in Table 5, many readers had read works
in more than one genre during the previous six months.
Classics, Historical, and Modern Novels
The survey also asked about the reading of classic works of
fiction, "historical novels," and "modern dramatic novels" that
did not fall into one of the genre categories. Classics had been read
by 19 percent of fiction readers, or about 7 percent of all adults.
Comparable figures for historical novels were 35 percent of fic-
tion readers, or 14 percent of adults, and for modern dramatic nov-
els, 31 percent of fiction readers, or 12 percent of adults. Of course,
the latter two categories encompass commercial bestsellers as well
as works with serious literary intentions.
30 Who Reads Literature?
TABLE 5. Proportions of U.S. Adult Population That Report Reading Various
Forms or Genres of Fiction Books in Last Six Months, U.S. Adults Aged 16
and Over, 1983.
Have Read Books of This Form or
Genre in the
Last Six Months
Percent of All
Percent of All
Literary Form
Fiction Readers
Adults (16+)
All Forms/Genres
100%
39%
Novels
Action/Adventure
37%
14%
Mystery/Detective
35%
14%
Historical
35%
14%
Modern Dramatic
31%
12%
Romance (Traditional)
28%
11%
Science Fiction
21%
8%
Spy/lnternat. Intrigue
19%
7%
Classics
19%
7%
Fantasy
17%
7%
Romance (Sexy)
13%
5%
Romance (Gothic/Hist.)
13%
5%
Occult/Supernatural
12%
5%
Westerns
10%
4%
War Books
10%
4%
Juvenile/Children's
26%
10%
Short Stories
22%
9%
Humor/Satire
20%
8%
Poetry
11%
4%
Plays
8%
3%
SOURCE: Market Facts, Inc. & Research & Forecasts, Inc. 1983 Consumer Re-
search Study On Reading And Book Purchasing. Vol. I: Focus On
Adults. New York: Book Industry Study Group, Inc., 1984.
What the Readers are Reading 31
Poetry, Short Stories, Drama
The BISG study found that 22 percent of fiction readers had
read a book of short stories in the previous six months. Eleven
percent had read one or more poetry books and 8 percent, one
or more books of plays. As a fraction of all respondents, the propor-
tions were about 9 percent for short stories, 4 percent for poetry,
and 3 percent for drama. The latter percentages are in reasonably
good agreement with those found in the ARTS survey to have read
works of literary merit, especially if the difference in reference
periods is taken into account.
Audience Size Reconsidered
The results summarized above indicate that literature experts are
correct when they say that the proportion of people who read fine
literature is far smaller than the 56 percent who report reading fic-
tion, poetry, or drama in the course of a year. If the SPPA esti-
mate of the number of literature readers were taken at face value,
it would mean that literature had a substantially larger audience
than most of the other arts. For example, the SPPA estimated that
some 95 million people read literature in 1985. This was over two-
and-a-half times more than the number projected to have visited
art museums (37 million), and over four times more than the esti-
mated number of people who attended classical music performances
(22 million). Indeed, the ostensible number of literature readers
was nearly as great as the 101 million who reported attending mo-
vies within a year. (Interestingly, the combined number of adult
trade books and mass market paperbacks sold yearly in the U.S.—
some 1.1 billion in 1985 — is about the same as the total number
of movie tickets sold annually.)36
What the ARTS and BISG findings show, however, is that many
of the professed literature readers read only genre fiction or sen-
timental verse, the literary equivalents of TV "shoot-em-ups" and
sitcoms, or "Top 40" popular music. The proportion who read
serious contemporary literature of all forms in the course of a year
seems to be about 7 to 12 percent of the adult population (the 12
percent figure coming from the proportion who reported they had
read "modern dramatic novels" in the BISG survey). This would
32 Who Reads Literature?
still make the audience for literature comparable to that for some
of the other arts, roughly the equivalent of the 16 million people
who attend jazz performances or the 20 million who see live dra-
ma each year.
At the same time, the size of the audience for literature could
be two-to-three times larger, depending on where one draws the
line between "entertainment" and "art." If one is prepared to take
seriously popular authors, such as horror-story writer Stephen King,
poet-illustrator Shel Silverstein, humorist Garrison Keillor, or mys-
tery writer John D. MacDonald, as at least some critics are, then
the public for literature might be more like a fifth to a quarter,
rather than a tenth, of the adult population. If, on the other hand,
one restricted the approved following to those familiar with excel-
lent but not widely known authors, such as poets Adrienne Rich
or James Merrill, then the size of the audience for contemporary
literature would become minuscule indeed.
A few points should be made here. First, it is difficult to make
a precise estimate of the overall size of the literary audience from
the ARTS and BISG studies, because their published reports do
not contain necessary summary tabulations, and because of am-
biguities and flaws in the coding and tabulation procedures used
in the studies. It would certainly be desirable to conduct a survey
that made more careful use of the follow-up questions developed
in these studies, with a larger sample and expert advice on the cod-
ing of various works and authors. Such a study, however, would
not resolve arguments over what is art and what is mere enter-
tainment.
Second, in attempting to gauge the size of the audience for
literature, it does not seem appropriate to limit the audience to those
who read serious contemporary works, any more than one would
wish to limit one's definition of the audience for classical music
to those who attend Steve Reich or Milton Babbitt concerts, or the
audience for visual art to those who come out for the latest exhibit
at the Hirshhorn or Guggenheim. In each of these publics, there
is a substantial segment of followers who stick with time-honored
works and are not terribly receptive to the new and challenging.
It hardly seems fair or wise to exclude these individuals from the
audience counts. Their skeptical judgments about the worth of con-
What the Readers are Reading 33
temporary writers, composers, and painters will, if past experience
is any guide, be supported in many instances by art historians of
the future. In other cases, of course, the new and sometimes diffi-
cult works of today will become part of tomorrow's established
canon.
Third, in estimating the size of the audience for poetry, the
distinction between those who read classic works only and those
who read contemporary as well as classic literature makes a sub-
stantial difference. If one includes those who read well-established
poetry, then the ARTS and BISG surveys indicate that the audience
for serious poetry is about six percent of the adult population. This
is larger than the sizes of the audiences for ballet or opera. On
the other hand, if one restricts the audience to those who read con-
temporary ''literary" poetry, then, as noted above, the poetry au-
dience amounts to one percent or less of the population.
Finally, looking at the empty rather than the full portion of
the glass, it is striking how many adults there are in the American
public who can read, are reasonably educated, and have been ex-
posed to at least some literature in the course of their schooling,
but who read nothing or virtually nothing in the way of fiction,
poetry, or drama on even an occasional basis. The 1985 SPPA found
that at least 44 percent of the adult population had not read a sin-
gle literary work in the course of a year. The majority of these
people— 62 percent— were high school graduates, and one in five
had some college education. Similarly, the BISG study found that
42 percent of the adult population were non-book readers, in the
sense that they had read newspapers or magazines, but not a sin-
gle fiction or non-fiction book during the previous six months. Un-
fortunately, as noted earlier, the non-book-reading segment of the
population appears to be growing.
Chapter 3
Factors That Affect
Literary Participation
There are, from the start, a number of demographic characteris-
tics that affect a person's level of participation in the literary arts.
Education
In the 1985 SPPA data, if someone had not completed high
school, the odds were about two-to-one that he or she had not read
a novel, short story, poem, or play in the last 12 months. If the
person had a high school diploma, then the chances became slightly
better than fifty-fifty. But if the person had completed one or more
years of college, the odds were three-to-one in favor of him or her
being a literature reader.
Obviously, education was not a perfect predictor of literary
participation. Some people with relatively little education were regu-
lar readers of fiction, poetry, or drama, whereas a significant
minority of those with college training did not ordinarily read any
works of literature. Nevertheless, of the basic background varia-
bles, education was the one most closely correlated with literature
reading.
36 Who Reads Literature?
Education was also associated with poetry reading and crea-
tive writing, but not as strongly. (Table 6.) The proportion of peo-
ple who had read or listened to poetry was more than twice as large
among the college educated as among those with less than a high
school education. And the proportion who had tried to do crea-
tive writing was five times greater. But even among those with
graduate degrees, only a minority had read any poetry, and an even
smaller minority had done any creative writing in the last 12 months.
A person's educational attainment tends to be associated with
other social characteristics, such as his or her income level and
ethnic background. Thus, when education was combined with these
and other factors in an equation, the unique contribution of edu-
cation to the prediction of literary participation was somewhat
diminished* But education still remained the premier predictor, sur-
passing income and race, as well as age, sex, and residence. It
was also the leading predictor of poetry reading and creative writing.
There are a number of reasons why education should be a good
predictor of literary participation. The more years of education a
person has had, the more likely it is that he or she has been ex-
posed to literature in school and has had instruction in its ap-
preciation.
In addition, years of educational attainment could be used as
a proxy measure for intelligence. More intelligent individuals are
more likely to be avid and adept readers, to recognize and enjoy
good writing, and to share the interests and concerns of those who
write literature. Educated persons are also more likely to be ex-
posed to reviews, magazine and newspaper articles, public televi-
sion and radio programs, and the recommendations of friends.
Finally, more educated persons may feel social pressure to read
works of literature in order to be able to converse knowledgeably
about them with colleagues and friends.
As noted earlier, the association between educational attain-
ment and literature reading, and the rising levels of general edu-
cation in the United States, would lead one to expect that the amount
of literature reading is increasing. But other influences can over-
*Results of the predictive equations, which made use of a technique called logis-
tic regression analysis, are shown in greater detail in the Technical Appendix.
Factors That Affect Literary Participation 37
TABLE 6. Relationship Between Education and Income Levels and Literature
Reading, Poetry Reading, Creative Writing, and Book or Magazine Reading
in Last 12 Months, U.S. Adults Aged 18 and Over, 1985.
Proportion of Popu
lation Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56.0%
18.6%
6.2%
85.6%
EDUCATION GROUPS
Some College
75.4%
28.1%
1 1 .5%
97.2%
High School Grad
53.4%
14.6%
3.8%
85.9%
Less than HS
32.6%
12.1%
2.0%
68.4%
INCOME GROUPS
$25K & over
66.5%
22.6%
8.0%
92.3%
$10— 25K
51.8%
19.6%
6.4%
85.3%
Under $ 1 OK
43.6%
1 4. 1 %
3.2%
72.3%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
38 Who Reads Literature?
ride the effects of education on social behavior and produce trends
that are different from the expected ones. Voting is a good exam-
ple of this. As with the propensity to read literature, the propensi-
ty to vote is positively correlated with educational attainment. But
rising education levels have not resulted in increased levels of vot-
er turnout, at least not in recent decades. Moreover, as critics of
the educational system are quick to point out, the rise in general
education levels has been accompanied by some decay in educa-
tional quality. A high school diploma does not necessarily mean
as much as it once did in terms of skills mastered and knowledge
gained.
Income
Like education, an individual's income level is significantly
associated with literary participation. Among persons in the 1985
SPPA who had annual incomes of $25,000 or more, the odds were
about two-to-one that they had read a work of fiction, poetry, or
drama in the previous twelve months. For those with incomes be-
tween $10,000 and $25,000, the odds dropped to just over fifty-
fifty. And among those with incomes below $10,000, the chances
were about six-to-four against their being literature readers.
Income level was also correlated with the general reading of
books and magazines, and with the reading of poetry and creative
writing. (Table 6.) However, the relationships between income and
poetry reading, and income and creative writing, were considera-
bly weaker than the relationship with overall literature reading. For
example, those with incomes of $25,000 and over were only about
one-and-a-half times more likely to have read poetry than those
with incomes below $10,000.
A person's income is associated with his or her education level
and ethnic group, so some of the correlation between income and
literary participation could be due to these factors, rather than to
income per se. When income was combined with the other demo-
graphic factors in a logistic regression equation * the amount of
predictive power contributed by income, over and above that provid-
*This equation allows a "better fit" of the data by fitting them into a curve rather
than a straight line.
Factors That Affect Literary Participation 39
ed by education, turned out to be slight. Income was still a sig-
nificant, though weak, predictor of overall literature reading, but
not of poetry reading or creative writing.
Thus, those with higher incomes are more likely to be litera-
ture readers than those with lower incomes, primarily because the
former tend to be more educated than the latter. The fact that they
also have more money to buy books and more leisure time to en-
joy them may also play a role, but apparently not a major one.
Gender
Another basic characteristic that has a bearing on literary par-
ticipation is a person's gender. If a respondent in the 1985 SPPA
was a woman, the odds were nearly two-to-one that she had read
a novel, short story, poem, or play in the previous 12 months. For
men, by contrast, the odds were less than fifty-fifty. Women were
also more likely to have read books and magazines in general, to
have read poetry, and to have done creative writing, though all of
these relationships were considerably weaker than the association
with literature reading. (Table 7.)
When the demographic variables were combined in predic-
tive equations, gender proved to be the second-strongest factor (after
education) in separating literature readers and poetry readers from
non-readers. It was the fourth-strongest factor (after education, age,
and non-metropolitan residence) in differentiating creative writers
from non-writers.
In the BISG survey on reading, women were found to be much
more likely than men to be frequent book readers. Gender was
also associated with the amount of reading done: women were more
likely to be readers of fiction, and men of non-fiction. Men were
more apt to be readers of newspapers and magazines, but not books.
As might be expected, certain genres of fiction, such as romances,
had a largely female following, whereas other genres, such as ac-
tion/adventure stories and science fiction, had readerships that were
predominantly male.37
It would seem that both cultural and biological factors are at
work in accounting for the gender differences in literary partici-
pation. As discussed later, there is evidence that girls get more
encouragement to read from their parents. But there is also evi-
40 Who Reads Literature?
TABLE 7. Relationship Between Gender and Age/Year of Birth and Litera-
ture Reading, Poetry Reading, Creative Writing, and Book or Magazine Reading
in Last 12 Months, U.S. Adults Aged 18 and Over, 1985.
Proportion of Popu
lation Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56.0%
18.6%
6.2%
85.6%
GENDER
Female
63.0%
21.5%
7.4%
88.3%
Male
48.1%
16.2%
4.9%
82.7%
AGE/BIRTH YEAR
Young (18-29)
56.8%
19.7%
8.5%
87.0%
Middle (30-49)
60.8%
20. 1 %
7.6%
88.6%
Older (50 +)
50.3%
17.4%
3.0%
81.5%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
Factors That Affect Literary Participation 41
dence of innate differences between the sexes in the development
of reading skills and interests. Studies of standardized reading tests
given to elementary- school children have found that, on the aver-
age, girls read earlier, better, and more than boys do. Girls do not
surpass boys in all verbal areas: boys do as well or even slightly
better on vocabulary tests. But girls excel on tests of reading profi-
ciency, and fewer girls encounter difficulties in learning to read.38
Girls also write letters earlier and express more positive attitudes
toward reading stories.39
For reasons that are not well understood, women lose much
of their advantage over men on reading tests by late adolescence
and young adulthood.40 Among college-bound high school stu-
dents, for example, men score slightly higher than women on the
verbal portion of the Scholastic Aptitude Test (SAT), including the
reading comprehension subtest. On the other hand, women do
slightly better on the Test of Standard Written English that is giv-
en as part of the SAT, as well as on the English Composition
Achievement Test. Young women in high school and college con-
tinue to do more reading than men, especially reading for pleas-
ure, and to know more about literature. Thus, nearly twice as many
women as men take the College Board Achievement Test in Liter-
ature, and the mean score attained by women is significantly higher
than that for men.41
It might also be argued that women may be drawn to litera-
ture because of a greater interest in human character development
and social interaction patterns. In the past, women were raised in
a manner that called for sensitivity to other people's feelings and
motivations, and for getting one's way through persuasion rather
than assertiveness. Obviously, much of literature is concerned
with how people behave in various situations and why they act as
they do.
It is interesting to speculate about what effects the women's
movement has had and will have on female involvement with liter-
ature. Certainly, the drive for women's rights has helped to draw
attention to outstanding women writers, and to open more oppor-
tunities for women in the publication and promotion of literature.
One would also think, given the changes that women as a group
have been undergoing, that many would want to read or write about
42 Who Reads Literature?
their experiences and feelings in fictional, poetic, or dramatic forms.
Whenever norms and values are in flux, literature has a spe-
cial role to play. Literature can be a vehicle for exploring new pat-
terns of behavior and interaction. It can provide fictional characters
that serve as role models to real people going through similar strug-
gles. And it can give voice to both the exhilaration and the frus-
trations that many pioneers experience. Although many of the
best-known feminist authors, such as Betty Friedan and Germaine
Greer, are non-fiction writers, feminist issues and themes appear
in a broad range of contemporary fiction, including the works of
writers as disparate as Mary Gordon, Erica Jong, and Francine
du Pies six Gray.
Even the emergence of a new type of popular romance novels
with a more overtly sexual content can be at least partly attributed
to the women's movement, in the sense that the movement has made
it easier for women to be open about their sexuality. However, as
more women become involved in traditionally male career paths,
one wonders whether their reading patterns will become more like
the instrumental, non-fiction oriented reading of men.
Age
The year in which a person was born has relevance to literary
participation, both because it represents where the individual is
in his or her life cycle and because it indicates the historical peri-
od in which the person was raised. If literary participation pat-
terns are changing over time, the change should be reflected in
differences between age groups. The problem is in disentangling
historical change from aging effects. This is not completely possi-
ble with data from a single point in time, or even from two closely
spaced surveys. Some reasonable inferences can usually be drawn
about what is occurring, however, depending on the pattern of
change actually observed.
The wide range of birth years represented in the 1985 SPPA
was broken down into three broad groups: young adults (ages 18-29,
or birth years 1956-1967); middle-aged adults (ages 30-49, or birth
years 1936-1955); and older adults (ages 50 and older, or birth years
1935 and earlier). When this division was made, a relatively weak
relationship was found between age and literary participation: par-
Factors That Affect Literary Participation 43
ticipation declined from the middle to the older years. The propor-
tion reading literature, for example, decreased from 61 percent in
the middle years to 50 percent in the older years. Similar declines
were observed in creative writing, general reading, and poetry read-
ing, although the last difference was very slight. (Table 7.) Differ-
ences between the middle-aged and younger groups were so small
as not to be statistically significant, but were generally in the direc-
tion of the middle-aged reading more than young adults. Creative
writing was an exception, being higher in the young group, but
by very little.
Because education levels have been rising over time, age and
year of birth are correlated with educational attainment. Older
groups have lower education levels, on the average, than younger
age groups. Age and birth year are also somewhat correlated with
income levels (because middle-aged individuals tend to earn more
money than younger or older people) and with the sexual compo-
sition of the group (because women tend to live longer than men).
When education and other demographic variables were entered into
predictive equations along with age (which was treated as a con-
tinuous variable in the equations), the unique contribution of age
to the process of differentiating readers from non-readers was es-
sentially eliminated.
Thus, the decline in literature reading with age can be explained
by the correlation between birth year and education level. Older
people read less than younger ones, not because they are older (and
hence more infirm, or less energetic, or some such), but primari-
ly because they are less educated. This rinding has an important
implication for future consumption. Future cohorts of older Ameri-
cans, being more educated than the senior citizens of today, will
presumably be reading more literature. It may also be that the to-
tal volume of literature reading will increase, although the increase
in reading among the elderly may be offset by declines in reading
among young adults.
The apparent negative effect of age on literary participation
was not eliminated in the equation that differentiated creative writers
from non-writers. Although the effect of age was still quite weak,
it was the second-best predictor in the equation, after education.
This suggests that age as such has some debilitating or discourag-
44 Who Reads Literature?
ing effect on the production of imaginative writing. In her book
The Coming of Age, Simone de Beauvoir concludes that great age
is generally not conducive to literary creation, especially to the
writing of novels. She attributes this to the waning with age of the
"alacrity" and strength that imaginative writing requires. But de
Beauvoir also mentions notable exceptions to the rule, famous
authors like Sophocles, Cervantes, Voltaire, Victor Hugo, and Henry
James, who created some of their finest works in later life.42
The decline in amateur writing with age seems unfortunate
because older individuals, having experienced more, should have
more to write about. Once retired, they also have more time to
practice the craft of writing. Perhaps, as attitudes about what is
possible and appropriate for older people to do change, the de-
cline in writing associated with increased age will change as well.
Race/Ethnicity
Blacks and Hispanics are less likely to have read literature than
whites. The 1985 SPPA data show that the odds on someone who
was Black or Hispanic having read a novel, short story, poem, or
play in the previous 12 months were about 40-60 against. For non-
minority whites, on the other hand, the odds were nearly 60-40
in favor. In addition, whites were about 50 percent more likely than
Blacks or Hispanics to have read poetry or done some creative writ-
ing. (Table 8.) The rates for individuals from other minority groups
(predominantly Asians) generally fell between those of whites and
Blacks and Hispanics.
Educational Handicaps. Especially among older adults,
minority ethnic status is associated with lower educational attain-
ment and income levels in our society, despite the dramatic im-
provement in educational and employment opportunities for
minorities in the last three decades. Substantial fractions of Black
and Hispanic adults are either illiterate or "aliterate." Many
Hispanic-Americans and some Asian-Americans are literate in their
native languages, but not in English. A finding from the 1985 SPPA
illustrates these problems: one third of Blacks and Hispanics had
not read any kind of book or magazine in the last year. The com-
parable proportion among white adults was one tenth.
But when education, income, and other demographic factors
Factors That Affect literary Participation 45
TABLE 8. Relationship Between Ethnic Group Membership and Literature
Reading, Poetry Reading, Creative Writing, and Book or Magazine Reading
in Last 12 Months, U.S. Adults Aged 18 and Over, 1985.
Proportion
of Populate
Dn Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56.0%
18.6%
6.2%
85.6%
ETHNIC GROUPS
Whites
59.0%
20.0%
6.7%
89.9%
Blacks
43.0%
13.8%
4.5%
66.3%
Hispanics
41.5%
14.8%
4.0%
66.0%
Asians, Others
51.9%
16.0%
2.4%
85.3%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
46 Who Reads Literature?
were entered along with race into equations predicting literary par-
ticipation, the predictive power of race was considerably reduced.
(Because of the relatively small size of the Hispanic and Asian sub-
samples, only a variable differentiating Black from non-Black
respondents was entered into the predictive equations.) In the equa-
tions differentiating poetry readers from non-readers, and crea-
tive writers from non-writers, race added nothing to the prediction.
In the equation predicting overall literature reading, race remained
a significant, though weak, predictor. Similar results were obtained
in analyses with the data from the 1982 SPPA.
Socialization and skill differences. Minority individuals are
less likely to have been exposed to literature as children. Educa-
tional research studies have found that minority children, espe-
cially Hispanics, tend to have fewer reading materials in their homes
than non-minority youngsters, and are less apt to have been read
to by their parents.43 Consistent with this, Hispanic adults in the
SPPA reported that their parents generally had not encouraged them
to read books that were not required for school. In addition, the
quality of the formal education many minority individuals receive
is inferior to that received by the typical non-minority individual.
Thus, in the SPPA, fewer Black and Hispanic respondents report-
ed that they had been exposed to lessons in creative writing.
Furthermore, even though the basic reading skills and educa-
tional attainment levels of minority young people have risen sub-
stantially since the 1960s, standardized tests still show that the
reading proficiency of both Black and Hispanic youths lags behind
that of non-minority youths with equivalent years of education. In
1988, for example, the National Assessment of Educational Pro-
gress found that only about one-quarter of Black or Hispanic
17-year-olds could read on an adept level, whereas nearly half of
the white 17-year-olds were adept readers.44
Availability of minority literature. In addition to these educa-
tional barriers, there is the question of the availability of fiction,
poetry, and drama that is of interest to minority adults and reflects
their concerns and cultural traditions. The works of a few con-
temporary Black writers, such as Alex Haley, Toni Morrison,
Ntozake Shange, Alice Walker, and August Wilson, have received
widespread public attention in recent years. And some older Black
Factors That Affect Literary Participation 47
writers like Langston Hughes, James Baldwin, Ralph Ellison,
Richard Wright, and Lorraine Hansberry, have received recogni-
tion because of the enduring value of their work, and as a result
of "Black History Month" and other efforts to raise public con-
sciousness about the contributions of Blacks to American culture.
The sad truth, though, is that many Black young people are ig-
norant of these authors and their works. Moreover, although the
situation is far better than it was in the past, it could hardly be
said that there is as yet an extensive body of literary works by and
for Black Americans.
The situation is worse for Hispanic Americans. For one thing,
the Hispanic community is not a unified whole. It is divided into
Mexican-Americans, Puerto Ricans, Cuban-Americans, those from
Spain, and those from the different Central or South American
countries. Each of these groups has somewhat different traditions
and concerns. Most Hispanic-American authors are not well known
within their own communities and are virtually unknown to a broad-
er audience. Although there has been a surge of interest in Latin
American writers of late, this has had little carryover to Hispanic
authors writing in the U.S. Many of the latter continue to have dif-
ficulty getting their works published and disseminated to appropriate
audiences.
Residence
There is significant variation in literary participation across
different regions of the country and from urban to rural commu-
nities. These differences, however, are relatively modest and are
probably due mostly to differences in average educational level
across areas, or to the likely tendency of people who have literary
inclinations to prefer living in some areas over others.
Regional variations. In the data from the 1985 SPPA, the odds
that someone who lived in the West had read a novel, short story,
poem, or play in the last 12 months were almost two-to-one. By
contrast, the odds that someone from the South had done so were
only about 50-50. The odds for residents of the Northeast and Mid-
west were just slightly better than those for the nation as a whole.
(Table 9.) A similar pattern of regional variation was visible in the
data from the 1982 SPPA. Poetry reading and creative writing
48 Who Reads Literature?
TABLE 9. Relationship Between Region and Metropolitan Residence and Liter-
ature Reading, Poetry Reading, Creative Writing, and Book or Magazine Read-
ing in Last 12 Months, U.S. Adults Aged 18 and Over, 1985.
Proportion
of Popu
lation Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56.0%
18.6%
6.2%
85.6%
REGION
Northeast
57.0%
17.1%
4.6%
86.4%
Midwest
56.7%
21.0%
6.6%
90.3%
South
50.4%
18.5%
4.6%
80.6%
West
63.7%
17.0%
10.0%
87.2%
RESIDENCE
Central City
56.5%
18.5%
7.5%
85.5%
Suburbs
61.0%
21.2%
8.2%
91.2%
Non-Metro
48.9%
16.7%
2.8%
78.4%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
Factors That Affect Literary Participation 49
showed weaker and somewhat different patterns of regional varia-
tion, although the West was still the leading region as far as the
proportion doing creative writing was concerned. It was not pos-
sible to evaluate the predictive power of region after other factors
were controlled because the Census Bureau does not release both
geographic identifiers and household socioeconomic data in the
same public use files.
Urban-rural variations. The 1985 SPPA data showed that the
odds were about 60-40 that someone living in the suburbs of the
major metropolitan areas had read a work of literature in the previ-
ous 12 months. By contrast, the odds for a person living outside
of the metropolitan areas were less than 50-50. Residents of non-
metropolitan areas were also below average in rates of general read-
ing, creative writing, and poetry reading, although the differences
with respect to poetry were relatively slight. Residents of the cen-
tral cities of metropolitan areas were close to the national average
on each of the participation variables. Similar patterns of urban-
rural variation were found in the data from the 1982 SPPA.
The metropolitan residential factor was entered into predic-
tive equations by means of two variables, one identifying those who
lived in central cities, and the other, those who lived in non-
metropolitan areas. Only the latter added significantly to the predic-
tions. When education, income, age, and other demographic vari-
ables were taken into account, the contribution of non-metropolitan
residence was considerably reduced. Residence was, however, the
third-strongest predictor of creative writing (after education and
age). It was also a significant though weak predictor of overall liter-
ature reading. Thus, most of the negative effect of non-metropolitan
residence on literary participation is due to other characteristics
of the residents, such as their education levels and ages. There is,
however, some residual effect or correlate of residence that is not
accounted for by the demographic characteristics of the residents.
Predicting Participation from Demographics
In sum, the likelihood that a person will or will not be a read-
er of literature is significantly related to a number of basic back-
ground characteristics, the foremost being his or her education level.
While gender, age, ethnic background, income level, and place
SO Who Reads Literature?
of residence are also related, they tell only a limited amount about
the person's propensity to read. Other, more specific factors in the
individual's history and current life situation, such as parental en-
couragement to read, also come into play, and are examined in the
next section. But first, it is useful to see how well literary partici-
pation can be predicted when the basic background characteris-
tics are combined into predictive equations.
Literature reading. Five variables were entered into the equa-
tion for discriminating literature readers from non-readers. (In this
case, as in each of the later equations, differing numbers of varia-
bles were relevant and entered into the equation.) For the 1985
SPPA, education and gender were the predominant predictors, with
income, race, and non-metropolitan residence adding tiny but
statistically significant increments of predictive power. The equa-
tion was able to classify 68 percent of the survey respondents cor-
rectly. (Bear in mind that one would get about a 50 percent correct
classification by simply alternating between predictions of "read-
er" and "non-reader," and 56 percent correct by predicting that
everyone was a literature reader.) There was also a moderately good
correlation between the predicted probability of being a reader and
the actual response. The model did somewhat better at identifying
those who were readers (71 percent correct) than those who were
not (63 percent correct). An almost identical equation and similar
predictive results were obtained with the data from the 1982 SPPA.
Poetry reading. Only two variables — education and gender —
were entered into the equation for differentiating poetry readers
from non-readers. The equation classified 75 percent of the respon-
dents correctly, but given the relatively small proportion of poetry
readers in the survey, one would get about 80 percent correct by
predicting that no one had read a poem. Of course, the latter strategy
would lead to a complete misidentification of those who actually
did read poetry (a zero "hit rate"). On the other hand, the equa-
tion correctly identified 35 percent of those who had read poetry
and 83 percent of those who had not. The rank-order correlation
between predicted probability and response (r = .32) was moder-
ate, but weaker than that obtained with the literature reading equa-
tion. The equation and predictive accuracy obtained with the 1982
data were similar, although the additional (but weak) predictors
Factors That Affect Literary Participation SI
of age and non-metropolitan residence figured into the 1982
equation.
Creative writing. Four variables were entered into the equa-
tion for discriminating creative writers from non-writers. Once
again, education was the leading predictor, but this time age was
the second-best predictor. Non-metropolitan residence and gen-
der also figured into the equation. The equation classified 92 per-
cent of the respondents correctly, about the same overall proportion
correct that one would get by predicting that no one had done any
creative writing in the last 12 months. However, the equation was
able to identify correctly 21 percent of the actual writers, as well
as 95 percent of the non-writers. The rank-order correlation be-
tween predicted probability and actual response (r = .54) was
moderately good. The predictive accuracy obtained with the 1982
SPPA data was nearly identical, and the equation similar, although
central city residence (rather than non-metropolitan residence) and
income figured into the 1982 equation.
Socialization and Training
Early Encouragement of Reading
One factor that markedly increases an adult's chances of be-
ing a regular reader of literature is having grown up in a family
where reading was practiced and encouraged. Studies of academ-
ic achievement in children consistently find that the parents' edu-
cation level and the academic orientation of the home are among
the best predictors of how well a child will do in school.45
Aspects of the home environment that correlate with achievement
include the number of books and other reading materials in the
home, whether the child was read to regularly, and whether the
parents encouraged the child to read books not required for school.
Similarly, the SPPA has found that one's participation in the arts
as an adult is correlated with the education level of one's parents
and with recollections of having been exposed to the arts by one's
parents when one was a child.46 Of the various relationships be-
tween childhood socialization indicators and measures of adult arts
participation that are covered in the survey, those involving par-
ticipation in literature are among the strongest.
32 Who Reads Literature?
Parents ' education level. The SPPA asked respondents to re-
port the highest grade or year of regular school their fathers and
mothers had completed according to six categories ranging from
"7th grade or less" to "completed college (4+ years)." Although
17 percent of the respondents in the 1985 survey could not recall
their father's education level and 13 percent could not recall their
mother's, most were able to come up with at least an approxima-
tion. For the purpose of the analyses reported here, the higher of
the two education levels was used; if only one parent's education
level was known, it was used. The proportion of respondents whose
parents attained each education level is shown in Table 10.
Respondents with college-educated parents were considerably
more likely to be literature readers than those whose parents had
less than a high school education. If the respondent's parents were
college graduates, the odds on the person having read literature
in the past 12 months were about four- to-one. However, if the par-
ents had only an elementary school education, the odds were
reduced to less than 50-50. Parent education was also related to
the chances of having read poetry or done creative writing, though
not as strongly.
As might be expected, the relationships between the literary par-
ticipation measures and parent's education were not as strong as
those with the respondent's own educational attainment. This is
partly because there is less recall error in the measure of the respon-
dent's own education. But it is mainly because the respondent's
education is a better indicator of his or her intelligence and educa-
tional experiences. Of course, parent's education and own educa-
tion are significantly correlated. Parent's education was also related
to the respondent's year of birth (with respondents born in more
recent years having better educated parents) and ethnic group (with
Black and Hispanic respondents having less educated parents than
non-minority respondents).
Parental encouragement of reading. SPPA participants were
asked: "Did your parents — or other adult members of the
household— encourage you to read books which were not required
for school or religious studies: often, occasionally, or never?" Of
those in the 1985 SPPA, 37 percent reported that their parents en-
couraged them to read often; 29 percent were encouraged occa-
Factors That Affect Literary Participation S3
TABLE 10. Relationship Between Parent Education Level and Literature Read-
ing, Poetry Reading, Creative Writing, and Book or Magazine Reading in Last
12 Months, and Proportion of Adults with Parents at Each Education Level,
U.S. Adults Aged 18 and Over, 1982 and 1985.
Proportion
of Populati
on Group Who. . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
1 982 Data
ALL ADULTS
56.4%
19.8%
6.5%
84. 1 %
PARENT'S EDUCATION
College grad plus
81.0%
35.9%
13.0%
97.9%
Some college
76.0%
33.4%
12.0%
96.8%
High school grad.
64.9%
22.0%
9. 1 %
91.1%
Some high school
56.9%
20.0%
4.6%
82. 1 %
Grade school only
43.1%
14.4%
2.6%
Proportior
71.5%
l of Adults
with Parents at Each
1985 Data
56.0%
Education Level
ALL ADULTS
1985
1982
PARENT'S EDUCATION
College grad plus
78.3%
15.4%
15.0%
Some college
78.3%
12.2%
9.4%
High school grad.
61.7%
35.2%
34.2%
Some high school
50.8%
10.4%
1 1.7%
Grade school only
42.9%
26.7%
100.0%
29.7%
100.0%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions by N. Zill and M. Winglee from public use data files.
54 Who Reads Literature?
sionally; and 34 percent, never.
The relationship between parental encouragement to read and
adult literature reading was quite strong, stronger even than the
relationship between the respondent's education level and litera-
ture reading. For persons who were frequently encouraged to read
as children, the odds were nearly four-to-one that they had read
a novel, short story, poem, or play in the last 12 months. For those
who were never encouraged to read, on the other hand, the odds
were more than two-to-one against them having read literature in
the last year. Parental encouragement was also related to the chances
of having done other types of reading or creative writing, though
not as strongly. (Table 11.)
As would be expected, reports that the parents encouraged the
respondent to read were related to the parents' education level. If
the parents were college graduates, 61 percent of the respondents
said they were often encouraged to read. On the other hand, if the
parents had an elementary education, only 25 percent were often
encouraged to read, and more than half were never encouraged.
Parental encouragement also varied across ethnic groups. It was
less common among Hispanics than among Blacks, whites, or Asi-
ans. Only 20 percent of Hispanic respondents reported that they
were often encouraged to read, and 54 percent said they were never
encouraged. (Table 12.)
Women were more likely than men to report that they had been
encouraged to read as children (42 percent of the women, as op-
posed to 32 percent of the men). Parental encouragement also varied
by year of birth, with those born more recently being considera-
bly more apt to have been encouraged as children. Only 26 per-
cent of those born in 1910 or earlier reported that they had often
been encouraged to read, and less than half had been encouraged
even occasionally. By contrast, 40 percent or more of those born
since World War II were given frequent encouragement, and 70
percent or more received at least occasional encouragement.
Limitations of the evidence of socialization effects. The data
just reported seem to provide evidence that the encouragement of
reading in childhood helps to form an abiding habit of reading for
pleasure and enlightenment. The differences across groups in paren-
tal encouragement are also generally consistent with the group
Factors That Affect Literary Participation 55
TABLE 1 1 . Relationship Between Socialization Factors (Parental Encourage-
ment of Reading, Respondent's Exposure to Creative Writing Lessons) and
Literature Reading, Poetry Reading, Creative Writing, and Book or Magazine
Reading in Last 12 Months, U.S. Adults Aged 18 and Over, 1982 and 1985.
Proportion of Popul.
ation Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
1985
1 Q°?
1 7UZ
ALL ADULTS 56.0%
19.8%
6.5%
84. 1 %
PARENTS ENCOURAGED READING
Often 79.0%
32.8%
10.4%
94.6%
Occasionally 57.0%
17.1%
6.0%
87.6%
Never 32.0%
9. 1 %
2.8%
64.8%
R HAD CREATIVE WRITING LESSONS
• *
Yes 88.2%
46.8%
25.2%
98.5%
No 49.6%
15.2%
2.7%
79.9%
*R denotes respondent.
SOURCE: National Endowment for the Arts, and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions by N. Zill and M. Winglee from public use data files.
56 Who Reads Literature?
TABLE 12. Frequency with Which Parents Encouraged Reading by Parent Edu-
cation Level, Year of Respondent's Birth, Ethnic Group, and Gender, U.S. Adults
Aged 18 and Over, 1985.
Parents Encourag
;ed Reading.
Often
Occasionally
Never
Total
Percent Distributions
ALL ADULTS
37.3%
29.0%
33.7%
100.0%
PARENT'S EDUCATION
College graduate
60.5%
28.0%
1 1 .5%
100.0%
Some college
52.7%
30.4%
16.8%
99.9%
High school grad.
40.7%
36. 1 %
23.2%
100.0%
Some high school
35.4%
32.8%
31.8%
100.0%
Grade school only
24.9%
23.6%
5 1 .4%
99.9%
YEAR OF R'S BIRTH'
1956-1967
40. 1 %
32.3%
27.6%
100.0%
1936-1955
38.6%
32.8%
28.6%
100.0%
1935 or earlier
33.6%
22.4%
44.0%
100.0%
ETHNIC GROUP
White
38.8%
29.7%
31.5%
100.0%
Black
37.9%
27.1%
34.9%
99.9%
Hispanic
20.2%
25.5%
54.3%
100.0%
Asian, other
43.6%
22.9%
33.5%
100.0%
GENDER
Female
42.3%
26.9%
30.8%
100.0%
Male
31.7%
31.3%
37.0%
100.0%
*R denotes respondent.
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
Factors That Affect Literary Participation 57
differences in literary participation that were reported earlier. A
few caveats are in order, however. To begin with, the evidence on
socialization effects is based on retrospective recall of parental edu-
cation levels and encouragement, rather than on observations or
reports made at the time. With such distant recall, there is the pos-
sibility that memory is distorting the past to make it consistent with
present behavior, or that reports of literature reading and parental
encouragement are related because of common response bias. Thus,
to be properly cautious, the evidence should really be seen as sug-
gestive rather than definitive.
Furthermore, even if the relationships between parental charac-
teristics and adult literary participation prove to be genuine, the
mechanism involved might be at least partly genetic, rather than
wholly environmental. The same criticism applies here as has been
applied to studies of family influences on children's school achieve-
ment.47 High parental education levels and encouragement of
reading could be seen as markers of high IQ, or of literary talent
and interest, which may be passed on to the child as much or more
through shared genes as through a nurturing home environment.
It should also be noted that while growing up in a home where
parents read a lot and reading materials are readily available is con-
ducive to later literary participation, it is not essential. In the past,
when educational opportunities were more limited, many individu-
als who became well-read adults were raised by parents who could
not or did not read themselves. It does seem possible for schools
and libraries to make up for what the home does not provide. On
the other hand, the findings on parental encouragement of reading
suggest that, in trying to teach young people to develop a lifelong
appreciation for literature, the emotional context in which the learn-
ing occurs is important.
Creative Writing Classes
In addition to family influences, adult reading habits are shaped
by the formal training a person has received. The SPPA found that
adults who had taken lessons in music, art, acting, ballet, or classes
in music or art appreciation, were more likely to attend or take
part in related artistic activities than people who had not taken les-
sons or classes.48 As described below, a similar relationship was
58 Who Reads Literature?
obtained between creative writing classes and literary participa-
tion. Here again, the issue arises of whether having taken a class
is a cause of later participation or merely an indicator that the person
has a predilection for the subject. Probably both mechanisms con-
tribute to the observed relationships.
Respondents in the arts surveys were asked whether they had
ever taken lessons or a class in creative writing. Those who said
they had were asked to specify in which of four age ranges (elemen-
tary school, secondary school, college, later adulthood) the class-
es were taken. In the 1985 SPPA, 18 percent of all adults said they
had taken creative writing lessons or classes at some point. Most
had received such instruction when they were of high school or
college age. (Table 13.) Only 3 percent had taken writing classes
when they were 25 or older. Practically identical proportions were
obtained in the 1982 SPPA.
Creative writing lessons were less common than music les-
sons (which had been taken by nearly half of all adults), crafts les-
sons (about a third had received these at some point), or visual
arts lessons (one quarter had taken these). They were about as fre-
quent as music appreciation or art appreciation classes, and more
common than acting or ballet lessons (each of which had been taken
by about one tenth of all respondents).
If the person had taken a lesson or class in creative writing,
the odds were nearly nine-to-one that he or she had read a novel,
short story, poem, or play in the last 12 months. For those who
had not taken such a class, the odds were about 50-50. Adults who
had taken writing classes were also more likely to have read po-
etry and books and magazines in general. (Table 11.) As might be
expected, there was a moderately strong relationship between tak-
ing writing classes and doing creative writing. Although only a
quarter of those who had ever taken a class in creative writing had
done such writing within the last year, this rate was eight times
higher than that for adults who had not taken such courses.
Significant correlations between writing instruction and liter-
ary participation were found no matter at what ages the writing
classes had been taken. However, courses taken in the college years
(18-24) seemed to make slightly more of a difference than those
at other ages.
Factors That Affect Literary Participation 59
TABLE 13. Number and Proportion of Adults Who Had Creative Writing
Lessons at Various Ages, U.S. Adults Aged 18 and Over, 1982 and 1985.
Nurr
iber
Pre
iportion
1985
1982
1985
1982
Age at Which
Lessons Were Taken
ALL AGES
30.6 mil.
29.7 mil.
18%
18%
Less than 12 yrs.
1.6
1.3
1%
1%
12-17 years
14.6
12.7
9%
8%
18-24 years
16.5
16.6
10%
10%
25 yrs. or more
5.0
5.0
3%
3%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions from public use data files.
60 Who Reads Literature?
The more education a person had, the more likely he or she
was to have taken a course in creative writing. Nearly 40 percent
of those with some college education had done so, as contrasted
to about 10 percent of those who stopped at high school, and only
3 percent of those who did not complete high school. Writing train-
ing was also more common among those with more educated par-
ents and parents who had encouraged reading. (Table 14.) The
chances of having had formal training in creative writing as part
of one's education have increased markedly in this century. Only
3 percent of those born in 1910 or earlier received such instruc-
tion, as opposed to about 15 percent of those born in the late 1930s
or early 1940s, and nearly 30 percent of those born since the
mid-1950s. Non-Hispanic white respondents were twice as likely
to have received some creative writing training as Black or Asian
respondents, and five times more likely than Hispanic respondents.
Women were slightly more likely than men to have taken such a
course.
Current Life Style
It seems plausible that people's literature reading habits are in-
fluenced by major aspects of their daily lives, such as their jobs,
marital situations, and family responsibilities. What people do for
a living shapes their interest, affects the amount of time and mon-
ey they have for reading and book purchasing, and exposes them
to other people who may encourage or discourage certain types
of reading. Similarly, a person's marital status and family situa-
tion have effects on interests, discretionary time and money, and
exposure to different types of people. Job, marital, and family cir-
cumstances also have a good deal to do with a person's need for
stimulation, solace, or escape.
As shown below, there were indeed associations in the arts
survey data between literature reading and aspects of daily life.
The associations proved to be weaker than one might expect, how-
ever, especially after controlling for related factors such as educa-
tion, income, age, and gender. These findings suggest that literature
reading is a fairly robust habit that can persist in the face of time
pressures and competition from other activities. The other side of
Factors That Affect Literary Participation 61
TABLE 14. Proportion of Adults Who Have Ever Taken Creative Writing Les-
sons by Respondent's Education Level, Year of Birth, Ethnic Group, Gender,
Parent's Education Level, and Parental Encouragement of Reading, U.S. Adults
Aged 18 and Over, 1985.
ALL ADULTS
EDUCATION LEVEL
Have Had Lessons In Creative Writing:
Yes
8.0%
No
Percent Distributions
82.0% I
Total
00.0%
Some college
38.7%
61.3%
100.0%
High school graduate
10.6%
89.4%
100.0%
Less than high school
2.6%
97.4%
100.0%
YEAR OF R'S BIRTH
1 956- 1 967
28.4%
71.6%
100.0%
1936-1955
20. 1 %
79.9%
100.0%
1935 or earlier
7.5%
92.5%
100.0%
ETHNIC GROUP
White
20.4%
79.6%
100.0%
Black
12.1%
87.9%
100.0%
Hispanic
4. 1 %
95.9%
100.0%
Asian, other
9.0%
91.0%
100.0%
GENDER
Female
19.0%
81.0%
100.0%
Male
16.9%
83.1%
100.0%
PARENT'S EDUCATION
College graduate
40.9%
59. 1 %
100.0%
Some college
36.2%
63.8%
100.0%
High school graduate
19.7%
80.3%
100.0%
Some high school
1 1.3%
88.7%
100.0%
Grade school only
5.1%
94.9%
100.0%
PARENTS ENCOURAGED READING
Often
32.7%
67.3%
100.0%
Occasionally
14.4%
85.6%
100.0%
Never
5.7%
94.3%
100.0%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by
N. Zill and M. Winglee from public use data files.
62 Who Reads Literature?
this coin is that those who are non-readers of literature do not sud-
denly take it up when placed in circumstances that would seem
to give them the opportunity to do so.
Employment and Student Status
The Survey of Public Participation in the Arts collected in-
formation about whether the respondent was currently employed,
and, if so, at what job and for how many hours per week. The
respondent's current employment status was significantly related
to all types of reading surveyed, as well as to creative writing. (Table
15.)
In general, those in the labor force (i.e. , those working or look-
ing for paid work) were more likely than those not in the labor
force to have read literature. Students were a notable exception to
this rule. They showed the highest rates of literary participation
of all the employment groups. For students in the 1985 SPPA, for
example, the odds were about three-to-one that they had read fic-
tion, poetry, or drama in the last 12 months. More than a third
had read poems and nearly a fifth had done some creative writing
in that period.
Of course, the high participation rates of students are partly
due to their being required to read works of literature for courses
they are taking. In addition, students tend to be immersed in the
world of books and to associate with others who read, recommend,
and talk about books. What many will find remarkable about the
SPPA findings, however, is not that students' reading rates are so
high, but that they are not higher.
Of men and women in the labor force, those who worked part-
time had somewhat higher rates of literary participation than those
who worked full-time. Those who had a job but were not at work
(because of illness, maternity leave, a labor dispute, etc.) also had
above-average rates of literature reading, but not of poetry reading
or writing. These differences support the notion that having more
non-work time available results in more reading of literature. How-
ever, people who work part-time are more likely to be female and
younger than those who work full-time. Thus, the factors of gen-
der and age contribute to the observed differences as well.
In contrast, those who were unemployed (i.e., without jobs
Factors That Affect Literary Participation 63
TABLE 15. Relationship Between Current Employment Status and Literature
Reading, Poetry Reading, Creative Writing, and Book or Magazine Reading
in Last 12 Months, U.S. Adults Aged 18 and Over, 1985.
Proportion of Popu
lation Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56.0%
18.6%
6.2%
85.6%
CURRENT EMPLOYMENT STATUS
In Labor Force
Working full time
56.0%
17.8%
7.0%
88.5%
Working part time
61.0%
29.2%
10.1%
90.0%
With job, not at work
64.5%
19.2%
4.8%
88.2%
Unemployed
50.5%
13.3%
3.9%
76.7%
Not In Labor Force
Student
74.5%
35.3%
19.2%
93.6%
Keeping house
55.4%
16.4%
3.2%
83.5%
Retired, other
49. 1 %
17.5%
2.5%
76.5%
Disabled
33.7%
14.4%
0.0%
67.2%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts, tabulations by N.
Zill and M. Winglee from public use data files.
64 Who Reads Literature?
and/or looking for work) showed below-average levels of literary
participation and reading in general. In this case, the factor of time
available to read was apparently negated by the generally lower
education and income levels, and higher concentrations of ethnic
minorities among the unemployed. Lower education levels were
also the dominant factor in the below-average reading and writing
rates shown by those who had retired from the labor force.
Those who were full-time homemakers had average rates of
literature reading, about the same as those who worked full-time
at paid jobs. Given that most of the homemakers were women,
however, the literary participation rates were lower than would be
expected. The demands of homemaking and childrearing may have
played a role here.
The small group that was not in the labor force because they
were disabled showed the lowest rates of literary participation. This
group had high proportions of older members with little educa-
tion and members of minority ethnic groups. In addition, some
of the people in this group had disabilities that made it difficult
or impossible for them to read.
Occupational Group
The type of occupation at which a person worked showed a
moderately strong relationship with literature reading. White col-
lar workers were generally above average in their reading habits,
whereas blue collar workers were below average. For those in
professional occupations, such as medicine, law, and college teach-
ing, for example, the odds were about three-to-one that they had
read a work of literature in the past 12 months. For sales and cler-
ical workers, the odds were about two-to-one. On the other hand,
for those in the skilled crafts, such as electricians, machinists,
mechanics, and tool and die makers, the odds were about six-to-
four against their having read literature. And for laborers, the odds
were two-to-one against. Service workers, such as waiters, barbers,
dental assistants, and flight attendants, were intermediate. The odds
that they had read some literature were slightly better than 50-50,
about the same as the national average. Similar relationships were
found with poetry reading and creative writing. (Table 16.)
Of course, a person's occupation is closely related to his or
Factors That Affect Literary Participation 63
TABLE 16. Relationship Between Occupational Class and Literature Reading,
Poetry Reading, Creative Writing, and Book or Magazine Reading in Last 12
Months, U.S. Adults Aged 18 and Over, 1985.
Proportion of Population Group Who
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56%
19%
6%
86%
Observed Proportions
OCCUPATIONAL CLASS
Professional
76%
34%
19%
98%
Managerial
71%
22%
1 1%
93%
Sales, Clerical
67%
22%
5%
94%
Service Workers
54%
21%
1 1%
86%
Craftsmen
42%
13%
3%
86%
Operatives
37%
9%
2%
68%
Laborers
36%
7%
Adjusted Propo
0%
rtions
81%
OCCUPATIONAL CLASS
Professional
60%
26%
14%
90%
Managerial
62%
16%
8%
87%
Sales, Clerical
60%
18%
3%
90%
Service Workers
57%
20%
10%
90%
Craftsmen
53%
18%
5%
90%
Operatives
48%
13%
4%
76%
Laborers
48%
1 1%
2%
88%
Note: Adjusted proportions derived through multiple classification analysis.
Proportions adjusted to compensate for variations across groups in age, sex,
education, income, ethnic composition, and other background characteristics.
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts. MCA analysis results
derived from: Robinson, John P., et al., Public Participation in the
Arts: Final Report on the 1985 Survey, College Park, MD: Universi-
ty of Maryland Survey Research Center, December 1986, Tables 3.3,
3.4, 5.3a & b, and 5.4a & b.
66 Who Reads Literature?
her educational attainment and income level. Thus, much of the
variation in reading habits across occupational classes could be
attributed to these factors, rather than to occupation per se. When
education, income, and other background factors were taken into
account, the differences among occupational classes were consider-
ably reduced. Some significant variation remained, though. The
adjusted odds were about six-to-four in favor of a person having
read literature if he or she were a professional, manager, or cleri-
cal employee, whereas they were slightly less than 50-50 if the
person were an operative (such as a truck driver) or a laborer.
Marital Status
At first glance, there seemed to be only a weak and some-
what inconsistent relationship between a person's marital situation
and his or her literature reading habits. Marital categories that con-
tained a predominance of younger persons, namely the never mar-
ried and separated, were slightly higher in literary participation,
whereas the widowed, a group comprising mostly older persons,
showed relatively low rates of reading and writing. The observed
differences, however, appeared to be more a matter of age and edu-
cation than of nuptial status. (Table 17.) After controlling for age,
education, and race, a small but interesting difference emerged:
people who were separated (but not those who were divorced) had
slightly higher rates of literature reading, poetry reading, and crea-
tive writing, than people in the other marital categories. These find-
ings suggest that people tend to turn to literature to help deal with
the personal crisis of marital separation.
Presence of Children
Taking care of children can be time consuming. Time use sur-
veys have shown that parents of young children, especially mothers,
spend less time in eating, sleeping, and non-child-related recrea-
tional activities than adults without children.49 In the 1985 SPPA
data, however, there seemed to be little difference between the liter-
ature reading habits of adults with children and those of adults with-
out children. After controlling for education, age, and other
demographic factors, a small but significant difference did emerge,
with parents of children under 6 years of age showing slightly lower
Factors That Affect Literary Participation 67
TABLE 17. Relationship Between Marital Status and Literature Reading, Po-
etry Reading, Creative Writing, and Book or Magazine Reading in Last 12
Months, U.S. Adults Aged 18 and Over, 1985.
Proportion of Popu
lation Group
Who . . .
Did
Read
Read
Read
Creative
Books,
Literature
Poetry
Writing
Magazines
ALL ADULTS
56%
19%
6%
86%
Observed Proportions
MARITAL STATUS
Never Married
57%
22%
1 1%
86%
Married
56%
18%
5%
87%
Separated
55%
27%
10%
84%
Divorced
57%
13%
6%
87%
Widowed
49%
15%
Adjusted
0%
Proportions
80%
MARITAL STATUS
Never Married
55%
19%
9%
83%
Married
56%
19%
6%
86%
Separated
60%
29%
10%
89%
Divorced
56%
14%
6%
89%
Widowed
57%
19%
4%
87%
Note: Adjusted proportions derived through multiple classification analysis.
Proportions adjusted to compensate for variations across groups in age, sex,
education, income, ethnic composition, and other background characteristics.
SOURCE: National Endowment for the Arts, and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts. MCA analysis results
derived from: Robinson, John P, et al., Public Participation in the
Arts: Final Report on the 1985 Survey, College Park, MD: Universi-
ty of Maryland Survey Research Center, December 1986, Tables 3.3,
3.4, 5.3a & b, and 5.4a & b
68 Who Reads Literature?
TABLE 18. Relationship Between Parental Status and Literature Reading, Po-
etry Reading, Creative Writing, and Book or Magazine Reading in Last 12
Months, U.S. Adults Aged 18 and Over, 1985.
Proportion of Population Group Who. . .
Did Read
Read Read Creative Books,
Literature Poetry Writing Magazines
ALL ADULTS 56% 19% 6% 86%
PRESENCE AND Observed Proportions
AGE OF CHILDREN
No children at home
One child under 6
Two children under 6
One child 6-1 I
Two children 6- 1 I
PRESENCE AND
AGE OF CHILDREN
No children at home
One child under 6
Two children under 6
One child 6- 1 I
Two children 6- 1 I
Note: Adjusted proportions derived through multiple classification analysis.
Proportions adjusted to compensate for variations across groups in age, sex,
education, income, ethnic composition, and other background characteristics.
For simplicity, groups with older children have been omitted.
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1985 Survey of Public Participation in the Arts. MCA analysis results
derived from: Robinson, John P, et al., Public Participation in the
Arts: Final Report on the 1985 Survey, College Park, MD: Universi-
ty of Maryland Survey Research Center, December 1986, Tables 3.3,
3.4, 5.3a & b, and 5.4a & b.
56%
19%
6%
85%
53%
15%
7%
90%
54%
18%
5%
87%
57%
17%
8%
84%
61%
20%
9%
92%
Adjusted
Proportions
57%
20%
6%
85%
50%
13%
5%
88%
51%
18%
5%
83%
55%
17%
8%
84%
57%
19%
10%
89%
Factors That Affect Literary Participation 69
rates of literature and poetry reading than parents of children 6
and older, or non-parents. (Table 18.) The differences might have
been greater if the survey had measured the number of books read,
rather than just the fact of having read literature or not.
Of course, for some adults, having children serves to bring
them back into contact with literature or to increase their reading,
at least of children's and youth-oriented books. In the BISG sur-
vey of book reading, more than a quarter of all adult fiction
readers— or 10 percent of all adults— had read a juvenile or chil-
dren's book in the last six months. Presumably much of this was
parents reading to young children or reading aloud with older chil-
dren. Reading to a child was also the third leading reason (after
reading for pleasure and general knowledge) that fiction readers
gave for reading. This reason was cited by 29 percent of the fic-
tion readers.50
The Role of Television
Television watching is often cited as an activity that competes
with reading and as a major reason why people do not read more
literature. Yet television can be a spur to purchasing books and
reading, as when an author appears on a talk show, a book is made
into a television program or movie, or is advertised on television
or mentioned or reviewed on a cultural program. In the BISG sur-
vey on book reading, respondents were asked to rate the impor-
tance of various factors in selecting books to read and purchase.
"Seeing a movie or TV show based on the book" was among the
top eight reasons for selecting a book, rated as "very important"
by more than a quarter of the readers, and at least "somewhat im-
portant" by 60 percent of them.51
Adults interviewed in the SPPA were asked to report the number
of hours they watched television on an average day. In the 1985
survey, close to 30 percent of all respondents reported that they
watched 4 or more hours per day, which is here categorized as a
"heavy" viewing pattern. About a quarter said they watched less
than 2 hours per day ("light" viewing). The remainder, about 45
percent, watched between 2 and 4 hours ("moderate" viewing).
A similar viewing breakdown was obtained in the 1982 SPPA. (Table
19.)
70 Who Reads Literature?
TABLE 19. Amounts of Daily Television Viewing Reported by U.S. Adults Aged
18 and Over, 1982 and 1985.
Percent Distribution
Estimated Number of
Viewers in Population
1985
1982
1985
1982
TELEVISION VIEWING
Light (<2 Hrs/Day)
25.6%
24.0%
43.5 mil.
39.3 mil.
Moderate (2-3 Hrs)
45.9%
44.8%
78.1
73.3
Heavy (4 Hrs plus)
28.5%
3 1 .2%
48.5
51.1
Total
100.0%
100.0%
170.1 mil.
163.7 mil.
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions by N. Zill and M. Winglee from public use data files.
TABLE 20. Relationship Between Television Viewing and Literature Reading,
Poetry Reading, Creative Writing, and Book or Magazine Reading in Last 12
Months, U.S. Adults Aged 18 and Over, 1982 and 1985.
Proportion of Population Group Who. . .
Did Read
Read Read Creative Books,
Literature Poetry Writing Magazines
1 982 Data
ALL ADULTS
56.4%
19.8%
6.5%
84. 1 %
TELEVISION VIEWING
Light (<2 Hrs/Day)
61.9%
28.8%
9.5%
81.2%
Moderate (2-3 Hrs)
58.8%
21.1%
6.7%
86.5%
Heavy (4 Hrs plus)
49.9%
14.6%
4.8%
79.3%
1985 Data
ALL ADULTS
56.0%
TELEVISION VIEWING
Light (<2 Hrs/Day)
59.6%
Moderate (2-3 Hrs)
56.4%
n.a.
n.a.
n.a.
Heavy (4 Hrs plus)
52.9%
SOURCE: National Endowment for the Arts and U.S. Bureau of the Census,
1982 and 1985 Surveys of Public Participation in the Arts, tabula-
tions by N. Zill and M. Winglee from public use data files.
Factors That Affect Literary Participation 71
When the reports of TV watching were cross-tabulated with
reports of literature reading, a negative but relatively weak rela-
tionship between reading and viewing emerged. In the 1985 data,
the odds that "light" TV viewers had read a work of literature in
the last 12 months were slightly better than average, about six-to-
four. For "heavy" viewers, on the other hand, the odds were slightly
below average, about 50-50. "Moderate" television viewers were
about average in their literature reading propensity.
A similar but slightly stronger relationship was obtained with
the 1982 survey data. (Table 20.) These data also permitted an anal-
ysis of the association between TV viewing and the other literary
participation measures, which was not possible with the 1985 sur-
vey. Both poetry reading and creative writing showed negative rela-
tionships with time watching television, with the relationship for
poetry being slightly stronger. Light TV viewers were twice as likely
to have read poetry or done some creative writing as heavy view-
ers. Interestingly, the relationship between TV viewing and the read-
ing of books and magazines was curvilinear, with the moderate
viewing group showing a slightly higher proportion of readers than
either the light or heavy viewing groups. This could be because
poorly educated non-readers are apt to be either heavy viewers of
television or non-viewers.
Countervailing tendencies. It may be that the overall associa-
tion between TV viewing and literature reading is not stronger be-
cause there are opposing tendencies at work. As noted earlier, those
who are active in one type of leisure activity tend to be active in
other types as well. Some people simply do more than others, even
though everyone is constrained by the number of hours in the day.
This phenomenon is recognized in the saying, "If you want some-
thing done, ask the busy person to do it." We also know that there
are large individual differences in reading speed. Moreover, time-
use studies tell us that television watching is often done as a secon-
dary activity; i.e., something that goes on while other activities
are occurring.52
At some level, however, there must be a trade-off between one
form of media participation and other forms. It seems likely that
the trade-off between television and literature reading would be
more visible if additional information about the types and quanti-
72 Who Reads Literature?
ty of reading done were available in the survey, or if the television
viewers were further subdivided, into selective and non-selective
viewers, for example.
Chapter 4
Expanding the Audience.
What Can Be Done?
The State of Literature Reading
The survey results reported here contain both good and bad news
for those who would like to see literature in America not only sur-
vive but flourish. The major piece of good news is that despite
concerns about illiteracy and aliteracy in the United States, more
than half of all American adults report that they have read some
fiction, poetry, or drama within the last year. Levels of reading
in the U.S. seem to be comparable to those in Great Britain and,
as far as can be determined, other industrialized countries. In ad-
dition, general education levels have risen, recent generations of
adults are more likely than older generations to have been en-
couraged to read as children, and growing numbers of people have
been exposed to creative writing classes.
The surveys indicate that older adults are less likely to be read-
ers of literature than middle-aged or young adults. However, the
differences in reading propensities appear to be more a function
of older citizens' lower education levels than of age per se, imply-
74 Who Reads Literature?
ing that literature reading levels among the elderly should go up
in the future as the current cohorts of elders are replaced by the
more educated senior citizens of tomorrow.
Other aspects of the survey results are less heartening. Follow-
up questions asking what people meant when they said they had
read novels or short stories revealed that some of the reports were
erroneous and most involved the reading of lightweight, genre fic-
tion (thrillers, romances, science fiction, horror stories, etc.) as
opposed to more significant and enduring works. Of the 56 per-
cent of adults who reported reading fiction, poetry, or drama within
a 12-month period, less than half had read works of literary merit,
comprising between a tenth and a quarter of the adult population.
Moreover, the audience for meritorious contemporary works ap-
peared to be smaller still, constituting something like 7 to 12 per-
cent of all adults. Thus, although most Americans can and do read,
followers of serious literature are distinctly in the minority.
Another discouraging finding is that while literature reading
is likely to increase among older Americans, it seems to be decreas-
ing among young adults. Data from several surveys point to a de-
cline during the 1970s and 1980s in the frequency of reading among
those under the age of 30. Literature has also become an art that
is neglected by men and dominated by women. As of the mid-1980s,
women made up nearly 60 percent of the readers, and almost two-
thirds of the would-be writers of literature.
Whereas women are overrepresented, ethnic minorities con-
tinue to be underrepresented in the audience for literature. Despite
the growing visibility and influence of Black and Hispanic writers,
less than 45 percent of Black or Hispanic adults reported reading
fiction, poetry, or drama. Their lower reading rates are largely at-
tributable to their lower average education levels. But even when
they have equivalent years of schooling, national testing programs
have found that Black and Hispanic youths are less adept readers
than non-minority young people. Blacks and Hispanic adults have
had less exposure to creative writing classes than white adults, and,
as children, Hispanics were less apt to have been encouraged to
read by their parents.
The survey finding that may be most disappointing, however,
is the simple fact that large numbers of American adults— 44
Expanding the Audience: What Can Be Done? 75
percent— do not read literature at all. Most of the non-readers of
literature know how to read. They have completed high school and
been exposed to at least some instruction in literature apprecia-
tion. Yet they read nothing in the way of fiction, poetry, or drama.
Why is it that literature in general and quality literature in particu-
lar are not read more widely? What can be done to encourage such
reading?
Why Quality Literature Is Not Read More Widely
Three broad explanations can be suggested for why literature of
merit is not read more widely: a shortage of readers who appreci-
ate good literature, a dearth of writers who can communicate to
a mass audience while maintaining high literary standards, and a
need for more resources and knowledge to be applied to the pro-
motion of literary works. Much attention has been paid of late to
developments relevant to the first category; i.e. , to changes in our
educational system and broader society that may be producing fewer
citizens who appreciate good literature and fine art. These develop-
ments are of legitimate concern to all who value the arts and hu-
manities. When it comes to recommending steps to increase the
audience for literature, however, the suggestions that seem most
feasible to carry out fall mainly in the third category.
Readers Who Don 't Appreciate
Is American society turning out fewer adults nowadays who
have the skills and inclination to appreciate serious literature? Com-
mentators on the U.S. cultural scene have pointed to a number of
social trends that may be having stultifying effects on the enjoy-
ment of literature, and on the appreciation of other arts and hu-
manities as well.
Educational deterioration. Many critics claim that the U.S.
educational system has deteriorated, and that high schools and col-
leges are doing a poor job of transmitting the Western cultural her-
itage to students. The schools have been accused of not teaching
the skills required to appreciate great literature and art, not giving
students a solid grounding in the classics, not nurturing a love for
language, not requiring memorization of great poetry and prose,
76 Who Reads Literature?
allowing students to get away with careless writing, and other fail-
ings.53 Research findings lend some support to these criticisms,
but the picture is more complex than usually portrayed.
As is now well known, the College Entrance Examination
Board's Scholastic Aptitude Tests and other nationwide testing pro-
grams gave evidence of significant deterioration in student knowl-
edge and proficiency during the late 1960s and 1970s. Not only
did average test scores go down but also fewer students displayed
high levels of achievement in either verbal or quantitative skills.
Test scores have recovered somewhat during the 1980s, but the
achievement levels of today's college-bound students are still sig-
nificantly lower than those of comparable students in the early
1960s.54
The National Assessment of Educational Progress (NAEP) has
found that today's high school students know relatively little about
modern American literature, even though most have received in-
struction in literature appreciation.55 Earlier assessments showed
that student attitudes about reading literature become progressively
more negative as one goes from elementary school to junior high
to high school students.56 In 1985, NAEP assessed the literacy
skills of young adults (ages 21-25) and found that 95 percent could
read and understand the printed word, but only a small percentage
could understand complex material. For example, only 9 percent
of the young adults could understand an unfamiliar and rather subtle
short poem by Emily Dickinson well enough to explain what the
poet was trying to express.57
There is other research evidence, however, that casts litera-
ture instruction in U.S. schools in a more favorable light. For one
thing, U.S. schools are now at least trying to educate minority stu-
dents who were written off in the past and are still relatively neglect-
ed by educational systems in other nations. NAEP and other testing
programs have shown that significant progress has been made during
the last two decades in raising the basic reading and writing skills
of Black and Hispanic students.58 In an international comparison
of literature education in ten countries, Alan Purves and his col-
leagues found that "The United States brings a higher proportion
of its age cohort farther along in reading than any other country
in the sample without the best students suffering."59 Overall, U.S.
Expanding the Audience: What Can Be Done? 77
students did not fare badly in international tests of literature achieve-
ment, although their achievement was not quite as good as that
of British students in some areas and the U.S. students displayed
more negative attitudes toward literature than students in other coun-
tries. Analyses of the international test results showed that home
background was at least as important as school factors in account-
ing for individual differences in literature achievement. The ana-
lyses also called into question some of the prescriptions that have
been made for improving literature instruction. It was found, for
example, that students who did not frequently have to recite litera-
ture from memory performed better than those who did.60
Evidence from the College Board Achievement Testing Pro-
gram indicates that the study of literature may be growing more
popular, and U.S. high schools seem to be holding their own in
teaching literature appreciation to the best students. The number
of students who took the Literature Achievement Test increased
by nearly 50 percent between 1980-1981 and 1985-1986 (going from
15,556 to 22,955 students), and the mean score on the test increased
slightly (from 516 to 524) over the same interval* However, only
a small and rather select fraction of college bound students take
the Literature Test. (In 1985-1986 there were more than 1.6 mil-
lion who took the Scholastic Aptitude Test, 191 thousand, the Eng-
lish Composition Test, and more than 40 thousand, the American
History Test.)61
Technological change and cultural decay. In contrast to those
who blame our educational system for failing to maintain interest
in literature, other observers point to profound cultural and tech-
nological changes that have occurred in our society and say it is
unfair to expect the schools to overcome the negative effects of these
developments.62 Among the trends that may be working to the
detriment of literature appreciation are:
• the increased availability of alternative forms of entertain-
ment, not only television and movies, but also newspapers
with a variety of feature articles, specialty magazines, elec-
tronic games and personal computers, music videos, etc. ;
*Changes in test composition make it inadvisable to compare mean scores from
the 1980s with those from Literature Achievement Tests given in earlier years.
78 Who Reads Literature?
• the explosion of scientific and technical knowledge, which
has caused jargon to proliferate and compels the citizen
who wants to be reasonably well-informed to spend more
time reading factual material rather than literature;
• the breakdown of generally-accepted standards of artistic
quality and taste in the face of challenges by avant-garde
writers and artists, civil libertarians, ethnic minorities,
feminists, and others;63
• the emergence of a youth-oriented entertainment industry
that is blatantly vulgar and anti-intellectual, and that
produces and promotes rock music, movies, and television
shows aimed explicitly at the teenage and young adult au-
dience;64
• the advent of a so-called "lite era," in which the mass me-
dia and commercial advertising have trained viewers and
readers of all ages to be impatient with any work that re-
quires serious and sustained attention.
Although it certainly seems plausible that some or all of these
developments could have an effect on the reading and apprecia-
tion of literature, there has been no systematic research demon-
strating connections between these trends and changes in literary
participation.
The influence of television. Aside from the deterioration of the
educational system, the emergence of television as the dominant
medium of U.S. mass communication is most often cited as hav-
ing a degrading influence on American civilization. Television
programming has been described as addictive fare that is designed
primarily to keep viewers watching through the commercials, thus
taking up time that might otherwise be spent in reading or other
more constructive pursuits. Television has also been accused of
satiating the public appetite for narrative with "empty calories"
instead of intellectual substance, of reducing public taste to the
lowest common denominator, and of failing to challenge, inspire,
or enlighten the viewer. It could be argued as well that television
has lured writers who might produce works of broad and endur-
ing appeal away from serious writing and into more lucrative but
ephemeral projects, such as scripts for soap operas, situation come-
Expanding the Audience: What Can Be Done? 79
dies, and made-for-TV movies.
Only a weak negative association was found in the SPPA data
between television viewing and literature reading, but there is lit-
tle doubt that the advent of television has had profound effects on
our cultural life.65 Again, however, research that convincingly
demonstrates links between television and trends in literary par-
ticipation remains to be done.
Writers Who Don't Communicate
Some have argued that at least part of the blame for the rela-
tively small audiences that contemporary literature and art com-
mand must be laid at the feet of the writers and artists themselves.
The popular appeal of literature and the other arts has certainly
been affected by the separation of the serious writer, painter, or
composer from any sort of integral role in the operation or
ceremonial life of the society.
Just how far artistic alienation has come is illustrated by a re-
cent incident in which the late Robert Penn Warren, who was then
serving as poet laureate of the United States, expressed indigna-
tion at the suggestion that he might produce a poem or two on na-
tional or patriotic themes during his tenure as laureate.66 Instead
of feeling honored that he was being called on to be the poetic voice
of the nation, he apparently felt affronted by the notion. Warren
is certainly not alone in rejecting the role of people's spokesman.
Many contemporary writers and artists feel no obligation to deal
with themes that might be of concern and interest to large num-
bers of their fellow citizens, or to make their work understanda-
ble, let alone entertaining, to any but the cognoscenti. It is scarcely
surprising then, that the public chooses to stay away in droves from
the work of these writers and artists.
The current situation was eloquently summarized by publish-
er Dan Lacy in a 1980 talk at the Center for the Book in the Li-
brary of Congress. Lacy observed that:
The achievement of that communion between author and
reader, artist and viewer, composer and audience by which
creation is consummated depends on the possession of
a common vocabulary of words and forms and structures
80 Who Reads U terature?
of meaning. Over the years this common coin grows worn
with use so that the freshness and force of communica-
tion is blurred and dimmed. Young writers and painters
and composers yearn to shatter them for new forms that,
they feel, will better express their meaning. Better ex-
press indeed, but not better convey that meaning if the
new-minted forms are not part of the audience's curren-
cy. Communion fails, full creation is aborted, and the ar-
tist's work in whatever field becomes a solipsism, to which
he retreats with a greater willingness because of his grow-
ing contempt for and alienation from society.
One senses today how few are the artists in any field,
at any adequate level of competence, who feel the strong
central currents of society surge through them to shape
their work— in the sense that Shakespeare and Haydn and
da Vinci felt at one with their times. In another day even
those creators and thinkers who felt most alienated and
hostile to the dominant forces of their times— such as Karl
Marx, Zola, or Brecht— yet felt society itself important —
quite literally terribly important— and themselves and their
work important in challenging it. They were therefore
called forth to their utmost not only to express but to con-
vey their meanings, to reach minds, to engage themselves
to the fullest with the life of their time— whether as its
voice or its foe. I do not find it so today.67
Publishers Who Don 't Promote
In addition to the large-scale social changes described above,
there are more mundane reasons why contemporary literature is
not more widely read. These reasons have to do with a lack of
resources devoted to the promotion of literary books and deficien-
cies in their packaging, advertising, and distribution. In these areas,
there are actions available to private firms and public organiza-
tions that might help to boost the sales and readership of contem-
porary works of merit.
As things now stand, relatively little money or effort is spent
on publishing literary books, especially in comparison to the large
amounts spent promoting television programs, movies, popular
Expanding the Audience: What Can Be Done? 81
magazines, and other mass media products that compete with books
for the reader's attention. The modest resources that are invested
in promotion tend to be spent in standard ways: sending the au-
thor on a book tour, distributing free copies to reviewers and promi-
nent individuals who might provide testimonials, placing
advertisements in literary magazines or the book review sections
of newspapers, etc. Most of these methods consist largely of preach-
ing to the converted rather than trying to make new disciples from
among those who read only popular fiction, those who do not read
literature at all, or those from ethnic minorities and other social
groups who are underrepresented in the literary audience.
There is, to be sure, a good commercial reason why more pro-
motion is not done: the money to support it is not there. As men-
tioned earlier, most volumes of serious fiction, poetry, and drama
do not sell many copies, even if they have received excellent reviews.
Publishing these works is typically a losing or marginally profita-
ble proposition. More promotion might lead to more sales, but in
most cases the risks involved seem to be too great or the projected
sales too small to warrant the investment of additional resources.
Efforts to publicize literary works more widely could, of course, be
subsidized by the profits (if any) that publishers make on their more
successful books, or through promotional campaigns conducted
by libraries and booksellers, by cash and in-kind contributions from
corporations, and by grants from private foundations or public agen-
cies* All of these forms of subsidy are now customary in the per-
forming arts, and there seems little reason why they should not
be applied more widely to the art of literature. In addition to the
need for more resources devoted to promotion, however, promo-
tional efforts should be better informed by knowledge about why
people read and how they go about selecting the particular books
they do.
Applying Research to Encourage Literature Reading
A number of steps could be taken to apply research findings to
the process of disseminating information about new and classic
*The NEA's Literature Program does provide a small amount of support, on an
annual basis through matching grants, for "audience development projects". These
include literary promotion projects, small press bookfairs, radio programs, etc.
82 Who Reads Literature?
books. Illustrative suggestions are offered in the following para-
graphs.
Paying attention to subject matter. One research result that has
received insufficient attention from those who sell and lend books
is that one of the main reasons people choose to read the books
they do is because they are interested in the subject matter dealt
with in the books and are seeking to expand their general infor-
mation about the time, place, people, or events in question.68 This
finding applies to the reading of both fiction and non-fiction. Yet
most bookstores and libraries are organized as if the reasons for
reading fiction were entirely separate and distinct from those for
reading non-fiction. Fiction and non-fiction works are kept in differ-
ent areas and there is no easy way for someone who is interested
in, say, browsing through novels about the U.S. Civil War to do
so. A display or shelving system that brought together fiction and
non-fiction books on given topics might well tempt the person who
is interested in a subject, but who does not ordinarily read fiction,
to buy or borrow a novel that deals with the subject. Likewise,
in advertising a new work of fiction that deals with a given sub-
ject or period, publishers could make use of special interest peri-
odicals and mailing lists that would reach those with a proven
interest in the subject or period. At present, this is rarely done.
Guiding readers to books they are likely to enjoy. Book re-
search has shown that fiction readers could use more information
to help guide their selection of books to read. For example, a study
by Nicholas Spenceley and Peter Mann found that it was not un-
common for library patrons to borrow a novel just because it looked
interesting on the shelf, without prior knowledge of the author or
title. When they did this, however, they wound up having a posi-
tive reaction to the book only 40 percent of the time.69 This was
well below the satisfaction levels of readers who had more speci-
fic information about the title or author prior to borrowing a book.
This suggests that in order to increase the chances of reader satis-
faction, which would, in turn, lead to more reading of contem-
porary literature, librarians, publishers, and literature programs
should be providing potential readers with more guidance of the
following sort: "If you enjoyed (Book A), you're likely to enjoy
(Books B, C, and D)." Moreover, it would be preferable if this gui-
Expanding the Audience: What Can Be Done? 83
dance were based on actual surveys of reader satisfaction, rather
than on the judgment of individual experts or the desires of pub-
lishers to plug particular titles in their catalogs.
Getting genre fans to read quality fiction. It would appear that
more could be done to encourage the readers of genre fiction to
explore more serious literary works. One way of doing this is to
establish, through research, which works of quality literature are
apt to appeal to readers of a particular genre, and then to publicize
those works through advertisements and outlets that are likely to
reach the genre readers. Other steps that might be taken are to give
public recognition to those writers of thrillers, romances, science
fiction, etc., whose novels or short stories evince superior literary
qualities, and to encourage good writers who are not widely read
to attempt some genre or genre-like writing in order to build a bigger
following for their work.
Using newspapers to reach non-readers of literature. Surveys
show that one way to reach people who read but do not read liter-
ature is through newspapers and news magazines, suggesting that
more should be done to publicize new books and promote litera-
ture reading in general through newspapers* Books could also be
advertised more extensively in newspapers, and not just in the book
review sections. As is done for the performing arts, newspapers
might be persuaded to run a regular literary "billboard" that com-
bined small advertisements for a number of different books in one
section, with the advertising space being sold at reduced rates. Liter-
ature programs could also encourage newspapers to run more fea-
ture articles about books and authors, to bring back the serialization
of quality fiction in their pages, and to print more poems, particu-
larly ones that are relatively accessible to readers who have not
been steeped in Ezra Pound and Wallace Stevens.
Employing television more effectively. Increasing the amount
of television publicity for serious literature does not mean simply
getting more authors on talk shows, for authors' appearances do
not always enhance book sales. Valuable principles can be learn-
ed, though, from programs that have been successful at encourag-
*One attempt to do this is the PEN Syndicated Fiction Project, which has placed
short stories in major newspapers across the country since 1978.
84 Who Reads Literature?
ing reading and stimulating book sales. These include the children's
reading series Cover to Cover and Reading Rainbow, and the adult-
oriented book review program, Bookmark. Among the lessons these
shows teach are to select the books to be featured carefully, choos-
ing ones that have both high quality and wide appeal, to present
excerpts from the books' stories on the show, with illustrations or
dramatizations that help to involve the viewer, and to ensure that
the viewer can obtain the books without great difficulty. (The last
point includes making certain that the book is still in print.)
Supporting promising developments in book marketing. Liter-
ature support programs should also be making efforts to identify,
encourage, and disseminate information about promising innova-
tions in the marketing and distribution of literary books. Two re-
cent examples of developments that may make a difference in the
sales and readership of today's literature are the proliferation of
book discussion groups and the emergence of the trade paperback
series.
Book discussion groups are small gatherings of adults who
assign themselves a series of common readings and get together
regularly to discuss the books and socialize. These groups, which
have apparently become fairly popular in a number of metropoli-
tan areas, are a perfect mechanism for expanding the range of people
who read modern literature as well as the number of books read.
Libraries and publishers could help to suggest and supply reading
matter for these groups and stimulate the formation of more such
groups.
Trade paperback series, such as Vintage Contemporaries,
Scribners Signature Editions, and Penguin Contemporary Ameri-
can Fiction, are a group of original or reprinted novels by differ-
ent contemporary authors that are published in higher-priced
paperbound editions with an imprint name and a uniform cover
format. Books in the series also appear together in special book-
store displays, and these displays are often prominently exhibited
in both local literary bookstores and in chain stores. Novels pub-
lished in these series have sold 10-to-20 times as many copies as
the typical literary novel that comes out in an individual hard-cover
edition. Although some critics have qualms about books being
bought and sold by "brand name" rather than on their individual
Expanding the Audience: What Can Be Done? 85
merits, this marketing innovation seems to have given a number
of serious authors a substantial boost in readership.70
Promoting both established and developing authors. It might
seem logical to focus publicity efforts for contemporary literature
on authors whose works have artistic distinction but little hope of
commercial success. Promotional efforts for writers like John Barth,
Joan Didion, Joseph Heller, Anne Tyler, John Updike, Gore Vi-
dal, or Tom Wolfe seem unnecessary because their names are widely
known in literary circles and their books usually sell quite well
in comparison with most works of serious fiction. Yet, if the goal
is to expand the audience for contemporary literature beyond its
current bounds, promoting established as well as struggling authors
might well be in order. It is likely that the aforementioned writers
and their works are not familiar to most members of the public
at large, as demonstrated by the 1984 poll showing that most Ameri-
cans did not recognize the name George Orwell. Moreover, the
number of people who buy or borrow even a best-selling book by
one of these authors is small in comparison with the number of
college-educated adults in the U.S. or the number of people who
watch a prime-time television show. Thus, the notion of including
such prominent authors in literature promotion campaigns is far
from ridiculous. Indeed, their inclusion would seem to be a sensi-
ble way to get more people reading quality literature.
In conclusion, it seems possible that the readership of con-
temporary fiction, poetry, and drama could be greatly increased
if more private and public resources were devoted to the encourage-
ment of literature reading and if promotional efforts made better
use of research knowledge about why people read and how they
select the books that they do. Ways in which research findings could
be applied include paying more attention to the importance of sub-
ject matter in people's selections of books to read, providing poten-
tial buyers and borrowers with guidance about books they are likely
to enjoy, encouraging fans of genre fiction to explore more seri-
ous literary works, using newspapers to reach a wide array of readers
(including those who do not currently read literature), employing
television more effectively to promote books, supporting such
86 Who Reads Literature?
promising new developments as book discussion groups and trade
paperback series, and promoting works by established as well as
developing authors.
The actions suggested above will not work wonders. In the
long run, the viability and reach of the literary enterprise depends
less on marketing techniques than on how well our society can cul-
tivate readers with the skills and sensibilities to appreciate great
literature and writers with the craft and imagination to entertain,
challenge, and enlighten large numbers of their compatriots. Giv-
en the current situation, however, with many potential readers in
the population but few reading serious works on any but an occa-
sional basis, it does seem that increased investment in promotion-
al efforts would produce a notable and much needed expansion in
the audience for literature.
TECHNICAL APPENDIX
How well can we predict whether a person will be a literature reader,
knowing basic facts about him or her such as age, sex, race, edu-
cation, income, and place of residence? The statistical method used
to answer this question was logistic regression analysis. ! Like lin-
ear discriminant analysis, logistic regression finds a weighted com-
bination of characteristics that best accounts for the observed
distribution of people into two mutually exclusive classes (in this
case, readers and non-readers). Unlike linear models, however,
logistic regression fits the data to a curve or, in multiple dimen-
sions, a curved solid, rather than a straight line or rectilinear
solid.2
Specifically, let the dependent variable, Y, be equal to one if
the person is a reader, and zero if he or she is not. Then the prob-
ability, pi, that the ith individual is a reader, is represented by the
equation:
Pi = 1/{1 + e*-« -**»•«>]}
where e is the base of the natural logarithms, alpha is the intercept
term, and jSj is the regression weight for the jth predictor varia-
88 Who Reads Literature?
ble. The optimal values of the a and /3 weights are derived using
the modified Gauss-Newton method of maximum-likelihood esti-
mation. (The LOGIST computer program in the SAS statistical
software package was used to develop the models.)3
Logistic regression has several advantages over linear regres-
sion for predicting dichotomous outcomes like literature reading.
First, the logistic model is inherently interactive in its depiction
of the relationships among the predictor and criterion variable and,
as such, is probably closer to the underlying reality than is the
simple additive model of linear regression.4 Second, it is often
more difficult to achieve an increase in the probability of occur-
rence of an outcome at the extremes of its probability distribution
(i.e., when the probability is very low or very high). The logistic
model is able to accomodate such "floor and ceiling" effects. Linear
regression, by contrast, assumes that a unit change in the value
of the predictor variable will produce a constant level of change
regardless of where one is on the probability distribution of the
dependent variable.5
Third, logistic regression always yields predicted probabili-
ties between 0 and 1. Linear regression, on the other hand, can
produce predicted values beyond 0 and 1, in effect predicting prob-
abilities below zero and in excess of 100 percent. Finally, the logistic
model makes fewer assumptions about the underlying distributions
of variables (e.g. , no multivariate normality assumption for covar-
iates). When distributional assumptions are violated, logistic regres-
sion still yields unbiased estimates of the standard errors of
coefficients, whereas ordinary least squares regression may not.
Appendix Table I summarizes the results of multiple logistic
regressions performed on data from the 1982 and 1985 Survey of
Public Participation in the Arts. The dependent variable was whether
or not the respondent reported reading literature during the previ-
ous twelve months. The independent variables were the respon-
dent's age (in single years), sex (coded "1" if female, "0" if male),
race (coded "1" if black, "0" if non-black), educational attain-
ment (years of regular school completed), income (total dollars,
broken down into 14 categories), central city residence (coded "1"
if the respondent lived in the central city of a metropolitan area,
"0" otherwise), and non-metropolitan residence (coded "1" if the
Technical Appendix 89
respondent lived outside any metropolitan area, "0" if inside the
suburbs or central city of a metropolitan area). Independent models
were developed for the 1982 and 1985 surveys.
The top panel of the table shows the contribution of each demo-
graphic characteristic to the prediction of literature reading. The
middle panel gives the computed regression coefficients and their
standard errors. The bottom panel presents several measures of the
predictive accuracy of the model.
A chi-square test was performed to assess whether the logistic-
regression model gave a discernibly better prediction than a mod-
el based on the presumption of no association between the predic-
tors and the criterion. A statistic called R, derived from the model
chi-square, is one measure of the overall predictive ability of the
model. The R statistic is similar to the multiple correlation coeffi-
cient in the normal setting, and incorporates a correction for the
number of parameters being estimated.
Individual r statistics ("partial rs") were computed for each
predictor variable. Ranging in value from —1 to +1, the partial
r provides a measure of the contribution of each variable to the
prediction, net of the effects of the other predictors. In the top panel
of Appendix Table I, the independent variables are listed in rank
order, based on the relative sizes of their partial correlation coeffi-
cients. Rank correlation coefficients showing the unadjusted rela-
tionship between each predictor and the criterion are presented for
comparison.
Once the best-fitting logistic model has been determined, the
observed cases can be classsified as readers or non-readers based
on the model. A case is predicted to be 1 on the dependent varia-
ble if the estimated probability for that case is greater than a cho-
sen value. The proportion of cases correctly classified is another
measure of the predictive ability of the model presented in Table
I, as are the false positive rate and the false negative rate. The
former is the proportion of predicted positives (readers) who were
actually negatives (non-readers). The latter is the proportion of
predicted negatives (non-readers) who were actually positives
(readers).
Because the predicted probabilities are continuous, the point
at which they are divided into positives and negatives is somewhat
90 Who Reads Literature?
arbitrary. An alternative way of assessing the predictive ability of
the model, one which is independent of a specific cut-point, is to
calculate an index of rank-order correlation between the predicted
probabilities and the dependent variable. This measure is also shown
for each model.
Appendix Table II summarizes the results of multiple-logistic
regressions in which the dependent variable was whether the respon-
dent had read poetry or attended a poetry reading during the previ-
ous 12 months, as reported in the 1982 and 1985 SPPA. The
predictor variables were the same as those in Table I. Appendix
Table HI summarizes regression models in which the criterion was
whether the respondent reported doing any creative writing dur-
ing the previous 12 months, and the predictor variables were again
the same.
Substantive conclusions derived from these regression analyses
are described at appropriate points in the main text.
NOTES
1. S.H. Walker and D.B. Duncan, "Estimation of the Probability of
an Event as a Function of Several Independent Variables," Biometrika,
54 (1967): 167-179.
2. S.J. Press and S. Wilson, "Choosing Between Logistic Regression
and Discriminant Analysis," Journal of the American Statistical As-
sociation, 73 (1978): 699-705.
3. Frank E. Harrell, Jr., "The LOGIST Procedure," in SUGI Sup-
plemental Library User's Guide, 1983 Edition, (Cary, North Caro-
lina: SAS Institute, Inc., 1983), 181-202.
4. Alfred DeMaris, "Interpreting Logistic Regression Results: A Crit-
ical Commentary," Journal of Marriage and the Family, 52 (Febru-
ary 1990): 271-277.
5. S. Philip Morgan and Jay D. Teachman, "Logistic Regression:
Description, Examples, and Comparisons," Journal of Marriage and
the Family, 50 (November 1988): 929-936.
TABLES TO THE APPENDIX
92 Who Reads Literature?
TABLE I. Multiple Logistic Models Predicting Whether Adults Have Read Liter-
ature in Last Year, Based on their Demographic Characteristics (Age, Sex,
Race, Education, Income, and Metropolitan Residence), U.S. Adults 18 and
Over, 1982 and 1985.
Contribution of Indivi
dual Predictors
1982
1
985
Variable
Rank Obs. r Adj. r
Rank Adj. r
Education
1
.39
.27**
1
.26**
Gender
2
.14
i / * *
. 1 o
2
.16**
Income
3
.20
.05**
3
.06**
Race
4
-.10
.05**
4
-.05**
Age
5
-.11
.02**
—
.00
Non-Metro Residence
6
-.07
.01*
5
-.03*
Central City Residence
—
.00
.00
—
.00
Regression Coefficients
1982
1985
Standard
Standard
Variable
Beta
Error
Beta
Error
Intercept
-2.48
.109
-2.33
.105
Education
(years attained)
.13
.003
.12
.004
Sex (Male = 0,
Female = 1)
.87
.037
.86
.041
Income
.05
.007
.04
.006
Race (Non-Black = 0,
Black = 1)
- .45
.063
- .46
.069
Age (in years)
- .003
.001
.000
.001
Non-Metro Residence
(= 1, else = 0)
- .09
.043
- .18
.049
Central City Residence
(= 1, else = 0)
.02
.046
.003
.050
Predictive Accuracy
Index
1982
1985
Proportion of Cases
Correctly Classified
69%
68%
False Positive Rate
26%
27%
False Negative Rate
37%
38%
Multiple Correlation Coefficient (R)
.38**
.36**
Rank Correlation Between Predicted
Probability and Response
.49**
.48**
Number of Cases 1
5,667
12,361
Technical Appendix 93
** p <
.01
*p<
.05
+ p <
.10
Obs. r :
= c
Observed correlation between predictor and criterion.
Adj. r = Partial correlation between predictor and criterion, net of effects
of other predictors in model.
Source: N. Zill and M. Winglee, analysis of public use tapes from 1982 and
1985 Survey of Public Participation in the Arts, National Endowment
for the Arts, and U.S. Bureau of the Census, 1987.
94 Who Reads Literature?
TABLE II. Multiple Logistic Models Predicting Whether Adults Have Read Po-
etry in Last Year, Based on their Demographic Characteristics (Age, Gender,
Race, Education, Income, and Metropolitan Residence), U.S. Adults 18 and
Over, 1982 and 1985.
Contr
bution of Indiv
dual Predictors
l
982
1985
Variable
Rank Obs. r Adj. r
Rank
Adj. r
Education
1
27
.24**
1
.18**
Gender
2
08
,i r*
2
.09**
Age
3
-
1 1
.05**
—
.00
Non-Metro Residence
4
-
02
.03*
—
.00
Central City Residence
5
01
.02 +
—
.00
Income
6
07
.02 +
—
.00
Race
—
-
04
.01
3
-.03 +
Regression Coefficients
19
82
1985
Standard
Standard
Variable
Beta
Error
Beta
Error
Intercept
-3.02
.248
-2.89
.305
Education
(years attained)
.1 1
.007
.08
.009
Gender (Male = 0,
Female = 1)
.59
.088
.49
.1 15
Age (in years)
-
.009
.003
- .001
.003
Non-Metro Residence
(= 1, else = 0)
.21
.103
- .09
.137
Central City Residence
(= |, else = 0)
.18
.106
- .09
.139
Income
-
.03
.015
.009
.017
Race (Non-Black = 0,
Black = 1)
-
.23
.156
- .41
.216
Predictive Accuracy
Index
1982
1985
Proportion of Cases
Correctly Classified
76%
75%
False Positive Rate
68%
71%
False Negative Rate
13%
14%
Multiple Correlation Coefficienl
:(R)
29**
.21**
Rank Correlation Between Predicted
Probability and Response
41**
.32**
Number of Cases
3
,849
2,132
Technical Appendix 95
*• p < .01
* p < .05
+ p < .10
Obs. r = Observed correlation between predictor and criterion.
Adj. r = Partial correlation between predictor and criterion, net of effects
of other predictors in model.
Source: N. Zill and M. Winglee, analysis of public use tapes from 1982 and
1985 Survey of Public Participation in the Arts, National Endowment
for the Arts, and U.S. Bureau of the Census, 1987.
96 Who Reads Literature?
TABLE III. Multiple Logistic Models Predicting Whether Adults Have Done
Creative Writing in Last Year, Based on their Demographic Characteristics
(Age, Gender, Race, Education, Income, and Metropolitan Residence), U.S.
Adults 18 and Over, 1982 and 1985.
Contribution of Indivi
dual Predictors
1982
1985
Variable
Rank <
Dbs. r Adj. r
Rank
Adj. r
Education
1
.17
-J-)**
1
-J-}**
Age
2
-.14
1 A**
2
-.12**
Gender
3
.06
.10**
4
.08**
Central City Residence
4
.04
.05*
—
.00
Income
5
.03
.03 +
—
.00
Non-Metro Residence
—
-.04
.00
3
-.09**
Race
—
-.01
.00
—
.00
Regression Coefficients
1982
1985
Standard
Standard
Variable
Beta
Error
Beta
Error
Intercept
-3.54
.400
-3.93
.523
Education
(years attained)
.12
.012
.12
.017
Age (in years)
- .03
.005
- .03
.007
Gender (Male = 0,
Female = 1)
.65
.144
.55
.189
Central City Residence
(= 1, else = 0)
.41
.164
- .03
.208
Income
- .05
.024
.003
.028
Non-Metro Residence
(= 1, else = 0)
- .07
.179
- .84
.273
Race (Non-Black = 0,
Black = 1)
- .30
.253
- .31
.346
Predictive Accuracy
Index
1982
1985
Proportion of Cases
Correctly Classified
92%
92%
False Positive Rate
88%
87%
False Negative Rate
3%
3%
Multiple Correlation Coefficient (R)
-y~\ * *
.32**
Rank Correlation Between Predicted
Probability and Response
.51**
.54**
Number of Cases
3,848
2,137
Technical Appendix 97
** p <
.01
*p<
.05
+ p <
.10
Obs. r =
= C
Observed correlation between predictor and criterion.
Adj. r = Partial correlation between predictor and criterion, net of effects
of other predictors in model.
Source: N. Zill and M. Winglee, analysis of public use tapes from 1982 and
1985 Survey of Public Participation in the Arts, National Endowment
for the Arts, and U.S. Bureau of the Census, 1987.
Notes
1. Jonathan Yardley, "The Increasing Irrelevance of Writing," The Washing-
ton Post (Monday, December 8, 1986), D2. Idem, "Blanking Verse in
the L.A. Times," The Washington Post (Monday, May 11, 1987), B2.
2. Allan Bloom, The Closing of the American Mind (New York: Simon
and Schuster, 1987), 62.
3. E.D. Hirsch, Jr. , Cultural Literacy: What Every American Needs To Know
(Boston, Houghton Mifflin, 1987).
4. Nick Thimmesch, (Ed.), Aliteracy: People Who Can Read But Won't
(Washington, DC: The American Enterprise Institute, 1984).
5. Martha S. Hill, "Patterns of Time Use," in: Time, Goods, and Well-
Being, F. Thomas Juster and Frank P. Stafford, Eds. (Ann Arbor, MI:
Institute for Social Research, University of Michigan, 1985), 138.
6. John P. Robinson, et al. , Americans ' Participation in the Arts: A 1983-84
Arts-Related Trend Study, Final Report (College Park, MD: University
of Maryland Survey Research Center, 1986), 86-87.
7. National Assessment of Educational Progress, Literature and U.S. His-
tory: The Instructional Experience and Factual Knowledge of High
1 00 Who Reads Literature?
School Juniors , by A.N. Applebee, J. A. Langner, and Ina VS. Mullis
(Princeton, NJ: Educational Testing Service, October 1987), 10-13. See
also: Lynne V. Cheney, American Memory: A Report on the Humani-
ties in the Nation 's Public Schools, National Endowment for the Hu-
manities Report No. NEH-636 (Washington, DC: U.S. Government
Printing Office, 1987). Diane Ravitch and Chester E. Finn, Jr., What
Do Our 17-Year Olds Know? A Report on the First National Assessment
of History and Literature (New York: Harper & Row, 1987).
8. U.S. Bureau of the Census, Statistical Abstract of the United States: 1987
(107th edition), Table No. 368, "Quantity of Books Sold and Value of
U.S. Domestic Consumer Expenditures, by Type of Publication and Mar-
ket Area: 1974 to 1985." (Washington, DC: U.S. Government Printing
Office, 1986).
9. National Assessment of Educational Progress, Literature and U.S. His-
tory. Lynne V. Cheney, American Memory: A Report on the Humani-
ties in the Nation's Public Schools. Diane Ravitch and Chester E. Finn,
Jr., What Do Our 17-Year Olds Know?
10. J. P. Robinson, C. A. Keegan, T. Hanford, and T. A. Triplett, Public Par-
ticipation in the Arts: Final Report on the 1982 Survey, prepared under
Grant No. 12-4050-003 from the National Endowment for the Arts (Col-
lege Park, MD: University of Maryland Survey Research Center, Oc-
tober 1985), Chapter 7. See also R.J. Orend, Socialization And
Participation in the Arts, (National Endowment for the Arts, 1989).
11. National Assessment of Educational Progress, Music 1971-79: Results
from the Second National Music Assessment, Report No. 10-Mu-01 (Den-
ver, CO: Education Commission of the States, November 1981). Idem,
Art and Young Americans, 1974-79: Results from the Second National
Art Assessment, Report No. 10-A-01 (Denver, CO: Education Commis-
sion of the States, December 1981).
12. U.S. Bureau of the Census, Statistical Abstract of the United States: 1987
(107th edition), Table No. 370, "New Books and New Editions Published
by Subject, 1970 to 1985, . . . ," (Washington, DC: U.S. Government Print-
ing Office, 1986).
13. John P. Dessauer, "U.S. Retail Book Sales by Subject: A First Estimate,"
Book Research Quarterly, Vol. 2, No. 4, Winter 1986-1987, 15-17.
14. U.S. Bureau of the Census, Statistical Abstract of the United States: 1987
(107th edition), Table No. 370, "New Books and New Editions Published
by Subject, 1970 to 1985, . . .".
15. For methodology of the SPPA studies see Robinson et al., Public Par-
ticipation in the Arts: Final Report on the 1982 Survey. John P. Robin-
Motes 101
son, Carol A. Keegan, Marcia Karth and Timothy A. Triplett, Public
Participation in the Arts: Final Report on the 1985 Survey (College Park,
MD: University of Maryland Survey Research Center, December 1986),
Chapter 2 and Appendix A. For methodology of the ARTS survey see
John P. Robinson et al. , Americans ' Participation in the Arts: A 1983-84
Arts-Related Trend Study. For methodology of the BISG study see Market
Facts, Inc.and Research & Forecasts, Inc. , 1983 Consumer Research
Study on Reading and Book Purchasing. Vol.1: Focus on Adults (New
York: Book Industry Study Group, Inc., 1984), 219.
16. Peter Mann, From Author to Reader: A Social Study of Books (London:
Routledge & Kegan Paul, 1982), 147.
17. See Robinson et al. , Final Report on the 1982 Survey, Final Report on
the 1985 Survey.
18. The 65-percent figure represents the projected number of readers of liter-
ature divided by the projected number of readers of books and maga-
zines. When the two survey items were actually cross-tabulated against
one another, a small percentage of respondents were found to have giv-
en inconsistent responses. That is, they said that they read literature in
the previous 12 months, but did not read books or magazines. Thus,
the proportion derived from cross-tabulation is slightly lower than the
figure cited.
19. The BISG figures on fiction book readership (39 percent) and overall
book readership (50 percent) were lower than the SPPA figure on liter-
ature readership (56 percent). This is probably because the BISG used
a 6-month reference period, and the SPPA, a 12-month reference peri-
od. Also, the BISG questions referred only to books, whereas the SPPA
question included fiction, poetry, and plays in magazines as well as in
books. The BISG figure on overall readership (92 percent) was higher
than the SPPA figure (86 percent), probably because the BISG ques-
tion included newspapers, which were not mentioned by the SPPA.
20. Mann, From Author to Reader: A Social Study of Books, pp. 147-148.
Idem, "The Novel in British Society," Poetics, Vol. 12, 435-448, 1983,
442).
21. Kay Gill and Donald P. Boyden, (Eds.), Gale Directory of Publications,
120th Edition (Detroit: Gale Research Company, 1988). The Standard
Periodical Directory, Tenth Edition (New York: Oxbridge Communica-
tions, 1987). Yardley, "Blanking Verse in the L.A. Times."
22. Dessauer, "U.S. Retail Book Sales by Subject."
23. Market Facts, Inc., 1983 Consumer Research Study, 1: 167-168.
24. Dessauer, "U.S. Retail Book Sales by Subject."
1 02 Who Reads Li terature?
25. Market Facts, Inc., 1983 Consumer Research Study, 1: 71.
26. Robinson et al., Final Report on the 1982 Survey, Chapter 3.
27. L.A. Wood, "Demographics of Mass Market Consumers," Book Re-
search Quarterly, Vol. 3, No. 1, Spring 1987, 31-39.
28. Market Facts, Inc., 1983 Consumer Research Study, 1: 64-68.
29. U.S. Bureau of the Census, Statistical Abstract of the United States: 1987
(107th edition), Table No. 368, "Quantity of Books Sold and Value of
U.S. Domestic Consumer Expenditures, by Type of Publication and Mar-
ket Area: 1974 to 1985."
30. William S. Lofquist, "Book Publishing," U.S. Department of Commerce,
U.S. Industrial Outlook — 1988, 1988, 29-9.
31. J.G. Bachman, L.D. Johnston, and P.M. O'Malley, Monitoring the Fu-
ture (Ann Arbor, MI: University of Michigan, Institute for Social Re-
search, annual volumes, 1975-1986). See also: "Daily Activity Patterns
of High School Seniors," Select Committee on Children, Youth, and
Families, U.S. House of Representatives, U.S. Children and Their Fam-
ilies: Current Conditions and Recent Trends, 1989 (Washington, DC:
U.S. Government Printing Office, September 1989J; 226-227.
32. Lofquist, "Book Publishing," 29-7—29-11.
33. Robinson et al., Americans' Participation in the Arts.
34. Mann, "The Novel in British Society," 443.
35. Market Facts, Inc., 1983 Consumer Research Study, 1: passim.
36. U.S. Bureau of the Census, Statistical Abstract of the United States: 1987
(107th edition), Table No. 368, "Quantity of Books Sold and Value of
U.S. Domestic Consumer Expenditures, by Type of Publication and Mar-
ket Area: 1974 to 1985;" and Table No. 376, "Selected Recreational Ac-
tivities: 1970 to 1985."
37. Market Facts, Inc., 1983 Consumer Research Study, 1: passim.
38. Eleanor E. Maccoby and Carol N. Jacklin, The Psychology of Sex Differ-
ences (Stanford, CA: Stanford University Press, 1974), 83-87. National
Center for Health Statistics, K.W. Schaie and J. Roberts, "School
Achievement of Children 6-11 Years, as Measured by the Reading and
Arithmetic Subtests of the Wide Range Achievement Test, United States,"
Vital and Health Statistics, Series 11, No. 103, Public Health Service
(Washington, DC: U.S. Government Printing Office, June 1970). Idem,
"School Achievement of Children by Demographic and Socioeconom-
ic Factors, United States," Vital and Health Statistics, Series 11, No.
109, DHEW Publication No. (HSM) 72-1011, Public Health Service
Notes 103
(Washington, DC: U.S. Government Printing Office, November 1971).
National Center for Health Statistics, J. Roberts, "Intellectual Develop-
ment of Children as Measured by the Wechsler Intelligence Scale for
Children, United States," Vital and Health Statistics, Series 11, No. 107,
DHEW Publication No. (HSM) 72-1004, Public Health Service
(Washington, DC: U.S. Government Printing Office, August 1971). Idem,
"Intellectual Development of Children by Demographic and Socioeco-
nomic Factors, United States," Vital and Health Statistics, Series 11,
No. 110, DHEW Publication No. (HSM) 72-1012, Public Health Serv-
ice (Washington, DC: U.S. Government Printing Office, December 1971).
National Center for Health Statistics, D.K. Vogt, "Literacy Among
Youths 12-17 Years, United States," Vital and Health Statistics, Series
11, No. 131, DHEW Publication No. (HRA) 74-1613, Public Health Serv-
ice (Washington, DC: U.S. Government Printing Office, December
1973). National Assessment of Educational Progress, "Males Dominate
in Educational Success," Spotlight: NAEP Newsletter, Vol. VIII, No.
5, October 1975 (Denver, CO: Education Commission of the States).
Alan Feingold, "Cognitive Gender Differences Are Disappearing,"
American Psychologist, Vol. 43, No. 2, February 1988, 95-103.
39. Nicholas Zill and James L. Peterson, "Learning to Do Things Without
Help," in Luis M. Laosa & Irving E. Sigel (Eds.), Families As Learn-
ing Environments for Children (New York: Plenum Publishing, 1982),
343-374.
40. National Center for Health Statistics, D.C. Hitchcock and G.D. Pinder,
"Reading and Arithmetic Achievement Among Youths 12-17 Years, as
Measured by the Wide Range Achievement Test, United States," Vital
and Health Statistics, Series 11, No. 136, DHEW Publication No. (HRA)
74-1618, Public Health Service (Washington, DC: U.S. Government Print-
ing Office, February 1974).
41. L. Ramist and S. Arbeiter, Profiles, College-Bound Seniors, 1985 (New
York: College Entrance Examination Board, 1986), 12-13, 22-23, 102,
and 14, 24.
42. de Beauvoir, Simone, The Coming of Age, (translation by Patrick O'Brian
of La Vieillesse), (New York: G.P. Putnam's Sons, 1972), 398-404.
"Writing is. . .a complex activity: it means a simultaneous prefer-
ence for the imaginary and a desire to communicate. . .The pro-
ject of writing therefore implies a tension between a refusal of
the world in which men live and a certain appeal to men them-
selves: the writer is both for and against them. This is a difficult
position: it implies very lively passions; and to be maintained for
a considerable length of time it calls for strength.
"Old age reduces strength; it deadens emotion. . .The ten-
1 04 Who Reads Li ter a t lire?
sion born of the reconciliation of two projects that are if not con-
tradictory then at least divergent, slackens. The elderly writer finds
himself deprived of that quality which Flaubert called alacrite."
(pp. 400-401)
43. Edward C. Bryant, Ezra Glaser, Morris H. Hansen, and Arthur Kirsch,
Associations Between Educational Outcomes and Background Variables:
A Review of Selected Literature, Contract Report by Westat, Inc., A
Monograph of the National Assessment of Educational Progress (Den-
ver, CO: Education Commission of the States, 1974).
44. National Assessment of Educational Progress, The Reading Report Card,
1971-88: Trends from the Nation's Report Card, by Ina V.S. Mullis and
Lynn B. Jenkins, Report No. 19-R-01, (Princeton, NJ: Educational Test-
ing Service, January 1990), 64.
45. Edward C. Bryant, Ezra Glaser, Morris H. Hansen, and Arthur Kirsch,
Associations Between Educational Outcomes and Background Variables:
A Review of Selected Literature. Alison Clarke-Stewart, Child Care in
the Family: A Review of Research and Some Propositions for Policy (New
York: Academic Press, 1977). Lee Willerman, "Effects of Families on
Intellectual Development," American Psychologist, Vol. 34, No. 10, Oc-
tober 1979, 923-929.
46. Robinson et al., Final Report on the 1982 Survey, Chapter 7. Orend,
Socialization and Participation In The Arts.
47. Lee Willerman, "Effects of Families on Intellectual Development."
48. Robinson et al., Final Report on the 1982 Survey, Chapter 7. Orend,
Socialization and Participation In The Arts.
49. Susan G. Timmer, Jacquelynne Eccles, and Keith O'Brien, "How Chil-
dren Use Time;" and Martha S. Hill, "Patterns of Time Use," in F. Tho-
mas Juster and Frank P. Stafford (Eds.), Time, Goods, and Well-Being
(Ann Arbor, MI: Institute for Social Research, University of Michigan,
1985). John P. Robinson, How Americans Use Time: A Social-
Psychological Analysis of Everyday Behavior (New York: Praeger, 1977).
50. Market Facts, Inc., 1983 Consumer Research Study, 143-145.
51. Ibid., pp. 133-137.
52. Hill, "Patterns of Time Use," Table 7.1, 141.
53. William J. Bennett, American Education: Making It Work (Washington,
DC: U.S. Government Printing Office, April 1988). Bloom, Closing
of the American Mind, Cheney, American Memory, Hirsch, Cultural
Literacy. Elizabeth Castor, "The Day of the Laureate: Warren Assumes
His Mantle & Braves the Press, for Love of Poetry," The Washington
Hotes 105
Post, October 7, 1986, Dl. Ravitch, Finn, What Do Our 17-Year-Olds
Know?; The National Commission on Excellence in Education, A Na-
tion at Risk: The Imperative for Educational Reform (Washington, DC:
U.S. Government Printing Office, April 1983).
54. Bennett, American Education, 8-14. Congressional Budget Office, 1987,
Educational Achievement: Explanations and Implications of Recent
Trends (Washington, DC: Congressional Budget Office, 1987). Nicho-
las Zill and C.C. Rogers, "Recent Trends in the Well-Being of Children
in the United States and their Implications for Public Policy," in An-
drew Cherlin (Ed.), Family Change and Public Policy (Washington, DC:
The Urban Institute Press, 1988), Chapter 2.
55. National Assessment of Educational Progress, Literature and U.S.
History.
56. National Assessment of Educational Progress, Reading, Thinking, and
Writing: Results from the 1979-80 National Assessment of Reading and
Literature (Denver, CO: Education Commission of the States, October
1981).
57. National Assessment of Educational Progress, Literacy: Profiles of
America 's Young Adults, by Irwin S. Kirsch and Ann Jungeblut, Report
No. 16-PL-02 (Princeton, NJ: Educational Testing Service, 1986).
58. National Assessment of Educational Progress, The Reading Report Card,
1971-88: Trends from the Nation's Report Card, by Ina V.S. Mullis and
Lynn B. Jenkins, Report No. 19-R-01, (Princeton, NJ: Educational Test-
ing Service, January 1990). Idem., The Writing Report Card, 1984-88,
by Arthur N. Applebee, Judith A. Langer, Ina V.S. Mullis, and Lynn
B. Jenkins, Report No. 19-W-01, (Princeton, NJ: Educational Testing
Service, January 1990).
59. Alan C. Purves et al., Reading and Literature: American Achievement
in International Perspective, NCTE Research Report No. 20 (Urbana,
IL: National Council of Teachers of English, 1981).
60. Alan C. Purves, Literature Education in Ten Countries, International
Association for the Evaluation of Educational Achievement, International
Studies in Evaluation II (Stockholm: Almquist & Wiksell; New York:
John Wiley, 1973).
61. College Board Admissions Testing Program, Statistical Summaries for
Academic Years 1980-1981 through 1985-1986, unpublished tabulations
supplied by Alicia Schmitt, Educational Testing Service, (Princeton,
NJ, February 1987).
62. Yardley, "The Increasing Irrelevance of Writing."
1 06 Who Reads literature?
63. Hilton Kramer, The New Criterion Reader: The First Five Years (New
York: The Free Press, 1988). Jonathan Yardley, "Paradise Tossed: The
Fall of Literary Standards," The Washington Post, January 11, 1988, B2.
64. Robert Pattison, The Triumph of Vulgarity: Rock Music in the Mirror
of Romanticism (New York: Oxford University Press, 1987).
65. Rubenstein, E.A., G.A. Comstock, and J.P. Murray (Eds.), Television
and Social Behavior, (Vol. 4): Television in Day-to-Day Life: Patterns
of Use, (Washington, DC: US Government Printing Office, 1972).
Schramm, Wilbur, Big Media, Little Media: Tools and Technologies for
Instruction, (Beverly Hills, CA: Sage, 1977).
66. Heard, Alex, "Out There: Versed Things First," The Washington Post
Magazine, April 17, 1988, 9-10.
67. Dan Lacy, "Publishing Enters the Eighties," in The State of the Book
World, 1980, The Center for the Book Viewpoint Series, No. 4 (Washing-
ton, DC: The Library of Congress, 1980) 11-25; 23.
68. Market Facts, Inc., 1983 Consumer Research Study, 133-136; 145.
69. Nicholas Spenceley, "The Readership of Literary Fiction: A Survey of
Library Users in the Sheffield Area," M.A. Dissertation, Postgraduate
School of Librarianship, Sheffield University, 1980. As described in Peter
H. Mann, "The Novel in British Society."
70. Jonathan Yardley, "The Soft-Cover Salvation of Modern Fiction," The
Washington Post, February 16, 1987, D2.
Index
Index
A
Ancient Evenings, 24
age, as a factor determining reading habits, 42-44
Atlantic, The, 9
Arts-Related Trend Study (ARTS), 4, 10, 23, 25
B
Bellow, Saul, viii
Bestseller lists, 3
Bloom, Allan, 2
Book Industry Study Group (BISG), 4, 9, 10, 14, 18, 19
book sales, trends, 18
Brecht, Bertol, 26
C
Chekhov, Anton, 1
Children, as a factor determining reading habits, 66, 68
Catcher in the Rye, 2
Color Purple, The, 24
Colwin, Laurie, ix
Couples, 24
Closing of the American Mind, The, 2
creating writing, 27, 61
Cultural Literacy, 2
D
demographic characteristics of readers, 11; predicting participation
based on demographics, 49
Dessauer, John P., 9, 10
Dickens, Charles, 1, 3, 24
Dickinson, Emily, 1, 26
Dostoyevsky, 1
drama, 3
E
Eliot, T.S., 26
education, as a factor determining reading habits, 35-37, 52, 53;
deterioration, 75-77
Educational Testing Service, 3
employment, as a factor determining reading habits, 62-66
F
Family Circle, 25
fiction, see also novels, short stories, 29
Fitzgerald, F. Scott, 3
Frost, Robert, 26
G
gender, as a factor determining reading habits, 39-42
geography, distribution of readers, 12; variation in literary participa-
tion, 47-49
Gershwin, George, 3
Graham, Martha, 3
H
Hawthorne, Nathianel, 3
Hellman, Lillian, 26
Hemingway, Ernest, 1, 24
Hirsch, E.D., 2
Holt, Victoria, 24
I
income, as a factor determining reading habits, 38
Irvine, Washington, 1
J
James, Henry, 24
K
Keillor, Garrison, 32
King, Stephen, viii, 24, 32
Krantz, Judith, ix, 24
L
L'Amour, Louis, 24
leisure activities of readers, 14, 60
literature reading, past growth 17; decline 18; literary criticism, 3
Longfellow, Henry Wadsworth, 26
Los Angeles Times Book Review, 9
M
MacDonald, John D., 32
Mann, Peter, 25
Mailer, Norman, 24
marital status, as a factor determining reading habits, 66-67
mass market books, 31
Melville, Herman, 1
Merrill, James, viii, 32
minority writers, 74
movies, 31
N
nation of readers, 1
National Geographic, 25
Naylor, Gloria, ix
New Yorker, The, 9
Newsweek, 25
novels, reading, 24, 28; classical, historical, and modern, 29
O
O'Neil, Eugene, 1
Orwell, George, 2
P
Parade, 26
Parker, Charlie, 3
Percy, Walker, vii
Plain, Belva, ix
plays, reading of, 26, 28, 31
Poe, Edgar Allan, 1, 26
Pound, Ezra, 26
poetry, 3, 9; sale of, 9; reading of, 26, 28, 30
Pollock, Jackson, 3
publishers, promotion, 80-81
Pushkin, Alexander, 1
R
race/ethnicity, as a factor determining reading habits, 44-47; see also
minority writers
radio, see television
reader, typical profiled, 1, 6
Reader's Digest, 25, 25
reading, encouraging, 81-86
Redbook, 25
Rich, Adrienne, viii, 32
S
Sandburg, Carl, 26
Service, Robert \Y., 26
Shakespeare, William, 1, 3, 26
Shange, Ntozake, 26
Shaw, Bernard, 26
Sheldon, Sidney, 24
short story, 25, 28
Silverstein, Shel, 32
Sophie's Choice, 24
Soviet Union, 2
Steele, Danielle, viii
Streetcar Samed Desire, 2
Stoppars, Tom, 26
Styron, William, 24
Survey of Public Participation in the Arts (SPPA), 4, 7-22
T
television and radio, 2; role of, 69-72; influence of, 78-79
technological change, 77
Thoreau, Henry David, 1
Tolstoy, Alexei,l, 24
Twain, Mark, 1, 24
U
Updike, John, 24
United Kingdom, 25
W
Walker, Alice, 24
Whitman, Walt, 1
Welty, Eudora, viii
Williams, Tennessee, 2, 26
Williams, William Carlos, 26
women's movement, x
Y
Yardley, Jonathan, 1
young adults, reading habits, 19
$9.95
"A BOOK THAT SHOULD BE OF INTEREST TO
ANYONE IN THE LITERARY BUSINESS."
— Booklist
Nicholas Zill and Marianne Winglee present a revealing portrait of the nation's
reading habits. In addition to the well-publicized discovery that a frightening
number of Americans can't read, it turns out that many who can — don't.
Focusing on who reads serious modern American literature— works by such
writers as Ann Beattie, Gloria Naylor, David Mamet and Saul Bellow — the
authors draw on extensive demographic information to report that the reading
of contemporary works is the pastime of a distinct minority. According to Who
Reads Literature, the proportion of Americans who read serious literature is
between 10 and 25 percent; contemporary fiction, poetry, and drama attracts
at best between 7 and 12
percent of adults.
"The typical reader of
literature in the United
States today. . .is a middle
aged, white woman, living
in the midwestern suburbs,"
say the authors. And they
are more involved than non-
readers in activities that
traditionally are not
thought of as "cultural"
pursuits — watching tele-
vision, going to the movies,
playing sports, and doing
volunteer or activist work.
In addition to pinpointing
just who is and isn't reading ^"i^^^^^™^^^™""^^^""^^™^
Adrienne Rich, Eudora Welty, and John Updike, Zill and Winglee suggest ways
in which publishers, booksellers, and libraries might reach a wider audience by
changing their marketing and promotion strategies.
Who Reads Literature raises important questions about the vitality of our
culture and education system. It tells us more about the United States as a
nation of readers than simply what we have stacked up on our bedside table.
"The perceived image of the literary'
American reader as a bearded male
academic in a tweed jacket with leather
patches bears only scant connection
to reality. . . . It's a woman's world
now, and if we are to have in the next
generation a literature of any con-
sequence, it will be because women
make it so."
^From the foreword
by Jonathan Yardley*
* Adapted from Yardley's column "Readership Down" published
in The Washington Post 8/20/90. Used by permission.
NICHOLAS ZILL is a social psychologist and
executive director of Child Trends, a non-profit
research organization in Washington, D.C.
MARIANNE WINGLEE is a senior statistical analyst
at Decision Resource .Corporation in Washington, D.C.