To Read

National Endowment for the Arts
To Read or Not To Read

A Question of National Consequence
Research Report #47
Reseaich Repoit #47
Novembei 2007
National Endowment for the Arts
1100 Pennsylvania Avenue, NW
Washington, DC 20506-0001
Telephone: 202-682-5400
Pioduced by the Oce of Reseaich & Analysis
Sunil Iyengai, Diiectoi
Sta contiibutois: Saiah Sullivan, Bonnie Nichols, Tom Biadshaw,
and Kelli Rogowski
Special contiibutoi: Maik Baueilein
Editoiial and publication assistance by Don Ball
Designed by Beth Schleno Design
Fiont Covei Photo: Getty Images
Piinted in the United States of Ameiica
Library of Congress Cataloging-in-Publication Data
To iead oi not to iead : a question of national consequence.
p. cm. (Reseaich iepoit , #47)
Pioduced by the Oce of Reseaich & Analysis, National
Endowment foi the Aits, Sunil Iyengai, diiectoi, editoiial and
publication assistance by Don Ball.
1. Books and ieadingUnited States. 2.
LiteiatuieAppieciationUnited States. I. Iyengai, Sunil, 1973
II. Ball, Don, 1964 III. National Endowment foi the Aits.
Z1003.2.T6 2007
028.9dc22
2007042469
202-682-5496 Voice/TTY
(a device foi individuals who aie deaf oi heaiing-impaiied)
Individuals who do not use conventional piint mateiials
may contact the Aits Endowments Oce foi AccessAbility at
202-682-5532 to obtain this publication in an alteinate foimat.
is publication is available free of charge at www.arts.gov,
the Web site of the National Endowment for the Arts.
5 Chaiimans Pieface
7 Executive Summaiy
23 Intioduction
SECTION ONE
Youth Voluntary Reading Patterns
27 Chaptei One
33 Chaptei Two
37 Chaptei Tiee
46 Chaptei Foui
SECTION TWO
What the Declines Mean for Literacy
55 Chaptei Five
63 Chaptei Six
68 Chaptei Seven
SECTION THREE
Why More than Reading is at Risk
77 Chaptei Eight
86 Chaptei Nine
94 Conclusion
96 List of Chaits and Tables
Co:r:s
To Read or Not To Read 5
T
o Read or Not To Read gatheis and collates the best national data available to
piovide a ieliable and compiehensive oveiview of Ameiican ieading today.
While it incoipoiates some statistics fiom the National Endowment foi the
Aits 2004 iepoit, Reading at Risk, this new study contains vastly moie data fiom
numeious souices. Although most of this infoimation is publicly available, it has
nevei been assembled and analyzed as a whole. To oui knowledge, To Read or Not
To Read is the most complete and up-to-date iepoit of the nations ieading tiends
andpeihaps most impoitanttheii consideiable consequences.
To Read or Not To Read ielies on the most accuiate data available, which consists
of laige, national studies conducted on a iegulai basis by U.S. fedeial agencies, sup-
plemented by academic, foundation, and business suiveys. Reliable national statisti-
cal ieseaich is expensive and time-consuming to conduct, especially when it iequiies
accuiate measuiements of vaiious subgioups (age oi education level, foi example)
within the oveiall population. Likewise, such ieseaich demands foimidable iesouices
and a commitment fiom an oiganization to collect the data consistently ovei many
yeais, which is the only valid way to measuie both shoit and long-teim tiends. Few
oiganizations outside the fedeial goveinment can manage such a painstaking task.
By compaiison, most piivate-sectoi oi media suiveys involve quick and isolated polls
conducted with a minimal sample size.
When one assembles data fiom dispaiate souices, the iesults often piesent con-
tiadictions. Tis is not the case with To Read or Not To Read. Heie the iesults aie
staitling in theii consistency. All of the data combine to tell the same stoiy about
Ameiican ieading.
Te stoiy the data tell is simple, consistent, and alaiming. Although theie has been
measuiable piogiess in iecent yeais in ieading ability at the elementaiy school level,
all piogiess appeais to halt as childien entei theii teenage yeais. Teie is a geneial
decline in ieading among teenage and adult Ameiicans. Most alaiming, both ieading
ability and the habit of iegulai ieading have gieatly declined among college giaduates.
Tese negative tiends have moie than liteiaiy impoitance. As this iepoit makes cleai,
the declines have demonstiable social, economic, cultuial, and civic implications.
Howdoes one summaiize this distuibing stoiy? As Ameiicans, especially youngei
Ameiicans, iead less, they iead less well. Because they iead less well, they have lowei
levels of academic achievement. (Te shameful fact that neaily one-thiid of Ameii-
can teenageis diop out of school is deeply connected to declining liteiacy and ieading
compiehension.) With lowei levels of ieading and wiiting ability, people do less well
in the job maiket. Pooi ieading skills coiielate heavily with lack of employment,
lowei wages, and fewei oppoitunities foi advancement. Signicantly woise ieading
skills aie found among piisoneis than in the geneial adult population. And decient
ieadeis aie less likely to become active in civic and cultuial life, most notably in vol-
unteeiism and voting.
Stiictly undeistood, the data in this iepoit do not necessaiily show cause and
eect. Te statistics meiely indicate coiielations. Te habit of daily ieading, foi
instance, oveiwhelmingly coiielates with bettei ieading skills and highei academic
Pvrrcr
Photo by Vance Jacobs
6 To Read or Not To Read
achievement. On the othei hand, pooi ieading skills coiielate with lowei levels of
nancial and job success. At the iisk of being ciiticized by social scientists, I suggest
that since all the data demonstiate consistent and mostly lineai ielationships between
ieading and these positive iesultsand between pooi ieading and negative iesults
ieading has played a decisive factoi. Whethei oi not people iead, and indeed how
much and how often they iead, aects theii lives in ciucial ways.
All of the data suggest how poweifully ieading tiansfoims the lives of individu-
alswhatevei theii social ciicumstances. Regulai ieading not only boosts the likeli-
hood of an individuals academic and economic successfacts that aie not especially
suipiisingbut it also seems to awaken a peisons social and civic sense. Reading
coiielates with almost eveiy measuiement of positive peisonal and social behavioi
suiveyed. It is ieassuiing, though haidly amazing, that ieadeis attend moie conceits
and theatei than non-ieadeis, but it is suipiising that they exeicise moie and play
moie spoitsno mattei what theii educational level. Te cold statistics conim
something that most ieadeis knowbut have mostly been ieluctant to declaie as fact
books change lives foi the bettei.
Some people will inevitably ciiticize To Read or Not To Read as a negative iepoit
undeistating the good woiks of schools, colleges, libiaiies, and publisheis. Ceitainly,
the tiends iepoited heie aie negative. Teie is, alas, no factual case to suppoit geneial
giowth in ieading oi ieading compiehension in Ameiica. But theie is anothei way
of viewing this data that is haidly negative about ieading.
To Read or Not To Read conimswithout any seiious qualicationthe cential
impoitance of ieading foi a piospeious, fiee society. Te data heie demonstiate that
ieading is an iiieplaceable activity in developing pioductive and active adults as well
as healthy communities. Whatevei the benets of newei electionic media, they pio-
vide no measuiable substitute foi the intellectual and peisonal development initiated
and sustained by fiequent ieading.
To Read or Not To Read is not an elegy foi the bygone days of piint cultuie, but
instead is a call to actionnot only foi paients, teacheis, libiaiians, wiiteis, and pub-
lisheis, but also foi politicians, business leadeis, economists, and social activists. Te
geneial decline in ieading is not meiely a cultuial issue, though it has enoimous con-
sequences foi liteiatuie and the othei aits. It is a seiious national pioblem. If, at the
cuiient pace, Ameiica continues to lose the habit of iegulai ieading, the nation will
suei substantial economic, social, and civic setbacks.
As with Reading at Risk, we issue this iepoit not to dictate any specic iemedial
policies, but to initiate a seiious discussion. It is no longei ieasonable to debate
whethei the pioblem exists. It is now time to become moie committed to solving it
oi face the consequences. Te nation needs to focus moie attention and iesouices
on an activity both fundamental and iiieplaceable foi demociacy.
Dana Gioia
Chairman, National Endowment for the Arts
I
n 2004, the National Endowment foi the Aits published Reading at Risk: ASurvey
of Literary Reading in America. Tis detailed study showed that Ameiicans in
almost eveiy demogiaphic gioup weie ieading ction, poetiy, and diamaand
books in geneialat signicantly lowei iates than 10 oi 20 yeais eailiei. Te declines
weie steepest among young adults.
Moie iecent ndings attest to the diminished iole of voluntaiy ieading in Ameii-
can life. Tese newstatistics come fioma vaiiety of ieliable souices, including laige,
nationally iepiesentative studies conducted by othei fedeial agencies. Biought
togethei heie foi the ist time, the data piompt thiee unsettling conclusions:
Americans are spending less time reading.
Reading comprehension skills are eroding.
These declines have serious civic, social, cultural, andeconomic implications.
A. Arrvics Avr Rroio Lrss
Teens and young adults iead less often and foi shoitei amounts of time when com-
paied with othei age gioups and with Ameiicans of the past.
1. Young adults are reading fewer books in general.
Neaily half of all Ameiicans ages 18 to 24 iead no books foi pleasuie.
Te peicentage of 18- to 44-yeai-olds who iead a book fell 7 points fiom 1992
to 2002.
2. Reading is declining as an activity among teenagers.
Less than one-thiid of 13-yeai-olds aie daily ieadeis.
Te peicentage of 17-yeai-olds who iead nothing at all foi pleasuie has
doubled ovei a 20-yeai peiiod. Yet the amount they iead foi school oi home-
woik (15 oi fewei pages daily foi 62 of students) has stayed the same.
Exrcu:ivr Surrvv
Percentage of Young Americans Who Read a Book Not Required for Work or School
Age group 1992 2002 Change Rate of decline
1824 59% 52% -7 pp -12%
2534 64% 59% -5 pp -8%
3544 66% 59% -7 pp -11%
All adults (18 and over) 61% 57% -4 pp -7%
pp = percentage points
Source: National Endowment for the Arts
Voluntaiy ieading iates diminish fiom childhood to late adolescence.
3. College attendance no longer guarantees active reading habits.
Although ieading tiacks closely with education level, the peicentage of college
giaduates who iead liteiatuie has declined.
65 of college fieshmen iead foi pleasuie foi less than an houi pei week oi not
at all.
Te peicentage of non-ieadeis among these students has neaily doubled
climbing 18 points since they giaduated fiom high school.
Percentage of Students Reading for Fun
Age 13 Age 17
Reading frequency 1984 2004 Change 1984 2004 Change
Never or hardly ever read 8% 13% +5 pp 9% 19% +10 pp
Read almost every day 35% 30% -5 pp 31% 22% -9 pp
Source: U.S. Department of Education, National Center for Education Statistics
Percentage Who Read Almost Every Day for Fun
1984 1999 2004
9-year-olds 53% 54% 54%
13-year-olds 35% 28% 30%
17-year-olds 31% 25% 22%
Percentage Who Read a Book the Previous Day (Outside School or Work)
In 2004
For at least 5 minutes For at least 30 minutes
8- to 10-year-olds 63% 40%
11- to 14-year-olds 44% 27%
15- to 18-year-olds 34% 26%
Source: The Henry J. Kaiser Family Foundation, Generation M: Media in the Lives of 8-18 Year-Olds (#7251), 2005
Percentage of Literary Readers Among College Graduates
Change Rate of decline
1982 1992 2002 19822002 19822002
82% 75% 67% -15 pp -18%
By the time they become college seniois, one in thiee students iead nothing at
all foi pleasuie in a given week.
4. Teens andyoung adults spendless time reading thanpeople of other age groups.
Ameiicans between 15 and 34 yeais of age devote less leisuie time than oldei
age gioups to ieading anything at all.
15- to 24-yeai-olds spend only 710 minutes pei day on voluntaiy ieading
about 60 less time than the aveiage Ameiican.
10%
20%
30%
40%
50%
60%
70%
21%
39%
26% 26%
T
o
t
a
l
None
Less than 1 hour
As high school seniors
in 2004
As college freshmen
in 2005
Reading per week:
Percentage of U.S. College Freshmen Who Read Little or Nothing for Pleasure
10%
20%
30%
40%
50%
60%
70%
21%
35%
28% 28%
T
o
t
a
l
None
Less than 1 hour
As high school seniors
(mainly pre-2002)
As college seniors
in 2005
Reading per week:
Percentage of U.S. College Seniors Who Read Little or Nothing for Pleasure
Source: University of California, Los Angeles, Higher Education Research Institute
By contiast, 15- to 24-yeai-olds spend 2 to 2 houis pei day watching TV. Tis
activity consumes the most leisuie time foi men and women of all ages.
Liteiaiy ieading declined signicantly in a peiiod of iising Inteinet use. Fiom
19972003, home Inteinet use soaied 53 peicentage points among 18- to 24-
yeai-olds. By anothei estimate, the peicentage of 18- to 29-yeai-olds with a
home bioadband connection climbed 25 points fiom 2005 to 2007.
i
5. Even when reading does occur, it competes with other media. is multi-
tasking suggests less focused engagement with a text.
58 of middle and high school students use othei media while ieading.
Students iepoit using media duiing 35 of theii weekly ieading time.
20 of theii ieading time is shaied by TV-watching, videoicomputei game-
playing, instant messaging, e-mailing oi Web suing.
i
U.S. Census Buieau, Computer
and Internet Use in the United
States, 1997 and 2003, and
PewiInteinet & Ameiican Life
Pioject, Home Broadband
Adoption 2007.
Percentage Using Other Media While Reading
7th-12th Graders in 20032004
% who use other media while reading
Most of the time 28%
Some of the time 30%
Most/some 58%
Little of the time 26%
Never 16%
Little/never 42%
Source: The Henry J. Kaiser Family Foundation, Media Multitasking Among Youth: Prevalence, Predictors
and Pairings (# 7592), 2006
Average Time Spent Reading in 2006
Hours/minutes spent reading
Weekdays Weekends
and holidays
Total, 15 years and over :20 :26
15 to 24 years :07 :10
25 to 34 years :09 :11
35 to 44 years :12 :16
45 to 54 years :17 :24
55 to 64 years :30 :39
65 years and over :50 1:07
Source: U.S. Department of Labor, Bureau of Labor Statistics
Percentage of 18- to 24-Year-Olds Reading Literature
1982 1992 2002
Percentage reading literature 60% 53% 43%
Change from 1982 # -7 pp -17 pp
Rate of decline from 1982 # -12% -28%
6. American families are spending less on books than at almost any other time
in the past two decades.
Although nominal spending on books giew fiom 1985 to 2005, aveiage annual
household spending on books diopped 14 when adjusted foi ination.
ii
Ovei the same peiiod, spending on ieading mateiials dipped 7 peicentage
points as a shaie of aveiage household enteitainment spending.
Amid yeai-to-yeai uctuations, consumei book sales peaked at 1.6 billion
units sold in 2000. Fiom 2000 to 2006, howevei, they declined by 6, oi
100 million units.
iii
Te numbei of books in a home is a signicant piedictoi of academic
achievement.
ii
Foi the puipose of this analysis,
family oi household is used
instead of the Buieau of Laboi
Statistics technical teim con-
sumei unit. In addition to families
and households, a consumei unit
may desciibe a peison living
alone oi shaiing a household with
otheis oi living as a ioomei in a
piivate home oi lodging house oi
in peimanent living quaiteis in a
hotel oi motel, but who is nan-
cially independent.
iii
Albeit N. Gieco and Robeit M.
Whaiton, Book Industry TRENDS
2007 (New Yoik, N.Y.: Book
Industiy Study Gioup, 2007),
vaiious pages.
$26
$28
$30
$32
$34
$36
1985 1989 1993 1997 2001 2005
Average Annual Spending on Books, by Consumer Unit
Adjusted for Inflation
The Consumer Price Index, 19821984 (less food and energy), was used to adjust for inflation.
Percentage of Time Spent Reading While Using Other Media
7th- to 12th-Graders in 20032004
Percentage of reading time
Reading while:
Watching TV 11%
Listening to music 10%
Doing homework on the computer 3%
Playing videogames 3%
Playing computer games 2%
Using the computer (other) 2%
Instant messaging 2%
E-mailing 1%
Surfing websites 1%
Using any of the above media 35%
Source: The Henry J. Kaiser Family Foundation, Media Multitasking Among Youth: Prevalence, Predictors
and Pairings (# 7592), 2006
B. Arrvics Avr Rroio Lrss Wrii
As Ameiicans iead less, theii ieading skills woisen, especially among teenageis and
young males. By contiast, the aveiage ieading scoie of 9-yeai-olds has impioved.
1. Reading scores for 17-year-olds are down.
17-yeai-old aveiage ieading scoies began a slow downwaid tiend in 1992.
Foi moie than 30 yeais, this age gioup has failed to sustain impiovements in
ieading scoies.
Reading test scoies foi 9-yeai-oldswho show no declines in voluntaiy
ieadingaie at an all-time high.
Te dispaiity in ieading skills impiovement between 9-yeai-olds and 17-yeai-
olds may ieect bioadei dieiences in the academic and social climate of
those age gioups.
Average Test Scores by Number of Household Books, Grade 12 (20052006)
Average Average Average
science score civics score history score*
Reported number of
books at home
More than 100 161 167 305
26100 147 150 289
1125 132 134 275
010 122 123 265
* Science and civics scores range from 0 to 300. History scores range from 0 to 500.
10
-6
-4
-2
0
2
4
6
8
1988 1990 1992 1994 1996 1999 2004
Reported as differences from 1984 reading scores.
Age 17
Age 9
Trend in Average Reading Scores for Students Ages 17 and 9
Test years occurred at irregular intervals.
Trend analysis based on data from the U.S. Department of Education, National Center for Education Statistics.
2. Among high school seniors, the average score has declined for virtually all
levels of reading.
Little moie than one-thiid of high school seniois now iead piociently.
iv
Fiom 1992 to 2005, the aveiage scoie declined foi the bottom 90 of ieadeis.
Only foi the veiy best ieadeis of 2005, the scoie held steady.
Te ieading gap is widening between males and females.
iv
Foi 12th-giadeis, Piocient
coiiesponds with a ieading scoie
of 302 oi gieatei (out of 500).
Average 12th-Grade Reading Scores by Gender
1992 2005
Female 297 292
Male 287 279
Male-female gap -10 -13
Change in 12th-Grade Reading Scores, by Percentile: 1992 and 2005
Percentile 1992 2005 Change
90th 333 333 0
75th 315 313 -2
50th 294 288 -6
25th 271 262 -9
10th 249 235 -14
All score changes from 1992 are statistically significant.
Percentage of 12th-Graders Reading at or Above the Proficient Level
1992 2005 Change Rate of decline
40% 35% -5 pp -13%
3. Reading prociency rates are stagnant or declining in adults of both genders
and all education levels.
Te peicentage of men who iead at a Piocient level has declined. Foi women,
the shaie of Piocient ieadeis has stayed the same.
v
Aveiage ieading scoies have declined in adults of viitually all education levels.
vi
Even among college giaduates, ieading piociency has declined at a 2023
iate.
4. Reading for pleasure correlates strongly with academic achievement.
Voluntaiy ieadeis aie bettei ieadeis and wiiteis than non-ieadeis.
Childien and teenageis who iead foi pleasuie on a daily oi weekly basis scoie
bettei on ieading tests than infiequent ieadeis.
Fiequent ieadeis also scoie bettei on wiiting tests than non-ieadeis oi
infiequent ieadeis.
v
Foi adults, Piocient coiie-
sponds with a piose liteiacy scoie
of 340 oi gieatei (out of 500).
vi
Exceptions aie adults still in
high school and those with a GED
oi high school equivalency. In
both cases, scoie changes fiom
1992 to 2003 weie not statistically
signicant.
Percentage of Adults Proficient in Reading Prose, by Gender
Female 14% 14% 0 pp 0%
Male 16% 13% -3 pp -19%
Both genders 15% 13% -2 pp -13%
Average Prose Literacy Scores of Adults, by Highest Level of Educational
Attainment: 1992 and 2003
Education level: 1992 2003 Change
Less than/some high school 216 207 -9
High school graduate 268 262 -6
Vocational/trade/business school 278 268 -10
Some college 292 287 -5
Associates/2-year degree 306 298 -8
Bachelors degree 325 314 -11
Graduate study/degree 340 327 -13
Percentage of College Graduates Proficient in Reading Prose
Bachelors degree 40% 31% -9 pp -23%
Graduate study/degree 51% 41% -10 pp -20%
Almost every day
Once or twice a week
Once or twice a month
Never or hardly ever
302
292
285
274
Almost every day
165
154
149
136
Average Reading Scores by Frequency of Reading for Fun
Grade 12 in 2005
Average Writing Scores by Frequency of Reading for Fun
Grade 12 in 2002
Reading scores range from 0 to 500.
Writing scores range from 0 to 300.
C. Tnr Drciirs i Rroio Hvr Civic, Socii, o Ecooric
Irviic:ios
Advanced ieadeis acciue peisonal, piofessional, and social advantages. Decient
ieadeis iun highei iisks of failuie in all thiee aieas.
1. Employers nowrank reading and writing as top deciencies in newhires.
38 of employeis nd high school giaduates decient in ieading compiehen-
sion, while 63 iate this basic skill veiy impoitant.
Wiitten communications tops the list of applied skills found lacking in high
school and college giaduates alike.
One in ve U.S. woikeis iead at a lowei skill level than theii job iequiies.
vii
Remedial wiiting couises aie estimated to cost moie than $3.1 billion foi laige
coipoiate employeis and $221 million foi state employeis.
viii
vii
Statistics Canada and OECD,
Learning a Living: First Results of
the Adult Literacy and Life Skills
Survey, 2005, 145.
viii
Te National Commission on
Wiiting, Writing: A Ticket to
Workor a Ticket Out: A Survey of
Business Leaders, 2004, 29, and
Writing: A Powerful Message from
State Government, 2005, 32.
Percentage of Employers Who Rate High School Graduates as Deficient
in Basic Skills
Writing in English 72%
Foreign languages 62%
Mathematics 54%
History/geography 46%
Government/economics 46%
Science 45%
Reading comprehension 38%
Humanities/arts 31%
English language 21%
Source: The Conference Board, Are They Really Ready to Work?, 2006
Percentage of Employers Who Rate Job Entrants as Deficient in Applied Skills
High school graduates deficient in: College graduates deficient in:
Written communication 81% Written communication 28%
Leadership 73% Leadership 24%
Professionalism/work ethic 70% Professionalism/work ethic 19%
Critical thinking/problem solving 70% Creativity/innovation 17%
Lifelong learning/self direction 58% Lifelong learning/self-direction 14%
Rated Very Important by Employers
Percentage of employers who rate the following basic skills as very important for high school graduates:
Reading comprehension 63%
English language 62%
Writing in English 49%
Mathematics 30%
Foreign languages 11%
2. Good readers generally have more nancially rewarding jobs.
Moie than 60 of employed Piocient ieadeis have jobs in management, oi in
the business, nancial, piofessional, and ielated sectois.
Only 18 of Basic ieadeis aie employed in those elds.
Piocient ieadeis aie 2.5 times as likely as Basic ieadeis to be eaining $850 oi
moie a week.
3. Less advanced readers report fewer opportunities for career growth.
38 of Basic ieadeis said theii ieading level limited theii job piospects.
Te peicentage of Below-Basic ieadeis who iepoited this expeiience was 1.8
times gieatei.
Only 4 of Piocient ieadeis iepoited this expeiience.
Percentage of Full-Time Workers by Weekly Earnings and Reading Level in 2003
$850$1,149 $1,150$1,449 $1,450$1,949 $1,950 or more Total earning $850
or more
Proficient 20% 13% 13% 12% 58%
Basic 12% 5% 2% 4% 23%
Below Basic 7% 3% 1% 2% 13%
Percentage Employed in Management and Professional Occupations, by Reading
Level in 2003
Management, business Professional Total in either job
and financial and related category
Proficient 19% 42% 61%
Basic 8% 10% 18%
Below Basic 3% 4% 7%
Percentage of Adults Who Said Their Reading Skills Limited Their Job
Opportunities, by Reading Level in 2003
A little Some A lot Total
Proficient 2% 1% 1% 4%
Basic 14% 15% 9% 38%
Below Basic 13% 22% 35% 70%
4. Good readers play a crucial role in enriching our cultural and civic life.
Liteiaiy ieadeis aie moie than 3 times as likely as non-ieadeis to visit
museums, attend plays oi conceits, and cieate aitwoiks of theii own.
Tey aie also moie likely to play spoits, attend spoiting events, oi do outdooi
activities.
18- to 34-yeai-olds, whose ieading iates aie the lowest foi any adult age gioup
undei 65, show declines in cultuial and civic paiticipation.
ix
5. Good readers make good citizens.
Liteiaiy ieadeis aie moie than twice as likely as non-ieadeis to volunteei oi do
chaiity woik.
x
Adults who iead well aie moie likely to volunteei than Basic and Below-Basic
ieadeis.
ix
National Endowment foi the
Aits, e Arts and Civic Engage-
ment: Involved in Arts, Involved in
Life, 2006.
x
Ibid.
Percentage of Adults Who Volunteered, by Reading Level in 2003
Less than Once a week Total who
once a week or more volunteered
Basic 16% 15% 31%
Percentage of Literary Readers Who Volunteered in 2002
Literary readers Non-readers Gap between groups
43% 16% -27 pp
Participation Rates for Literary Readers in 2002
Visit art museums 43% 12% -31 pp
Attend plays or musicals 36% 10% -26 pp
Attend jazz or classical concerts 29% 9% -20 pp
Create photographs, paintings, or writings 32% 10% -22 pp
Attend sporting events 44% 27% -17 pp
Play sports 38% 24% -14 pp
Exercise 72% 40% -32 pp
Do outdoor activities 41% 22% -19 pp
84 of Piocient ieadeis voted in the 2000 piesidential election, compaied
with 53 of Below-Basic ieadeis.
6. Decient readers are far more likely than skilled readers to be high school
dropouts.
Half of Ameiicas Below-Basic ieadeis failed to complete high schoola
peicentage gain of 5 points since 1992.
One-thiid of ieadeis at the Basic level diopped out of high school.
Foi high school diopouts, the aveiage ieading scoie is 55 points lowei than foi
high school giaduatesand the gap has giown since 1992.
Tis fact is especially tioubling in light of iecent estimates that only 70 of
high school students eain a diploma on time.
xi
xi
Editoiial Piojects in Education,
Diplomas Count 2007: Ready for
What? Preparing Students for
College, Careers, and Life after
High School, Executive Summaiy.
Percentage of Adults Who Voted in the 2000 Presidential Election, by 2003
Reading Level
Proficient 84%
Basic 62%
Below Basic 53%
Percentage of Adults at or Below Basic Prose Reading Level Who Did Not
Complete High School: 1992, 2003
Prose reading level
Below Basic Basic
1992 2003 Change 1992 2003 Change
45% 50% +5 pp 38% 33% -5 pp
Average Prose Reading Scores for Adult High School Graduates and Those Who
Did Not Complete High School: 1992, 2003
Prose reading score
Highest level of education 1992 2003 Change
Gap between groups -52 -55
7. Decient readers are more likely thanskilledreaders tobe out of the workforce.
Moie than half of Below-Basic ieadeis aie not in the woikfoice.
44 of Basic ieadeis lack a full-time oi pait-time jobtwice the peicentage of
Piocient ieadeis in that categoiy.
8. Poor reading skills are endemic in the prison population.
56 of adult piisoneis iead at oi below the Basic level.
Adult piisoneis have an aveiage piose ieading scoie of 25718 points lowei
than non-piisoneis.
Only 3 of adult piisoneis iead at a Piocient level.
Low ieading scoies peisist in piisoneis neaiing the end of theii teim, when
they aie expected to ietuin to family, society, and a moie pioductive life.
xii
xii
U.S. Depaitment of Education,
National Centei foi Education
Statistics, Literacy Behind Bars:
Results from the 2003 National
Assessment of Adult Literacy
Prison Survey, 2007, 77.
Percentage of Adult Prisoners and Household Populations by 2003 Reading Level
Prose reading level Household Prison Gap
Below Basic 14% 16% *+2 pp
Basic 29% 40% +11 pp
Intermediate 44% 41% *-3 pp
Proficient 13% 3% -10 pp
* = not statistically significant
Percentage of Adults Employed Full-Time or Part-Time, by 2003 Reading Level
Proficient 78%
Basic 56%
Below Basic 45%
Conclusion
Self-iepoited data on individual behavioial patteins, combined with national test
scoies fiom the Depaitment of Education and othei souices, suggest thiee distinct
tiends: a histoiical decline in voluntaiy ieading iates among teenageis and young
adults, a giadual woisening of ieading skills among oldei teens, and declining pio-
ciency in adult ieadeis.
Te Depaitment of Educations extensive data on voluntaiy ieading patteins and
piose ieading scoies yield a fouith obseivation: fiequency of ieading foi pleasuie
coiielates stiongly with bettei test scoies in ieading and wiiting. Fiequent ieadeis
aie thus moie likely than infiequent oi non-ieadeis to demonstiate academic
achievement in those subjects.
Fiom the diveisity of data souices in this iepoit, othei themes emeige. Analyses
of voluntaiy ieading and ieading ability, and the social chaiacteiistics of advanced
and decient ieadeis, identify seveial disciepancies at a national level:
Less ieading foi pleasuie in late adolescence than in youngei age gioups
Declines in ieading test scoies among 17-yeai-olds and high school seniois in
contiast to youngei age gioups and lowei giade levels
Among high school seniois, a widei iift in the ieading scoies of advanced and
decient ieadeis
A male-female gap in ieading pioclivity and achievement levels
A shaip divide in the ieading skills of incaiceiated adults veisus non-piisoneis
Gieatei academic, piofessional, and civic benets associated with high levels of
leisuie ieading and ieading compiehension
Longitudinal studies aie needed to conimand monitoi the eects of these diei-
ences ovei time. Futuie ieseaich also could exploie factois such as income, ethnicity,
iegion, and iace, and howthey might altei the ielationship between voluntaiy iead-
ing, ieading test scoies, and othei outcomes. Ciitically, fuithei studies should weigh
the ielative eectiveness and costs and benets of piogiams to fostei lifelong ieading
and skills development. Foi instance, such ieseaich could tiace the eects of elec-
tionic media and scieen ieading on the development of ieadeis in eaily childhood.
Recent studies of Ameiican time-use and consumei expendituie patteins high-
light a seiies of choices luiking in the question To iead oi not to iead? Te futuie
of ieading iests on the daily decisions Ameiicans will continue to make when con-
fionted with an expanding menu of leisuie goods and activities. Te impoit of these
national ndings, howevei, is that ieading fiequently is a behavioi to be cultivated
with the same zeal as academic achievement, nancial oi job peifoimance, and global
competitiveness.
Technical Note
Tis iepoit piesents some of the most ieliable and cuiiently available statistics on
Ameiican ieading iates, liteiacy, and ieadei chaiacteiistics. No attempt has been
made to exploie methods foi ieading instiuction, oi to delve into iacial, ethnic, oi
income tiaits of voluntaiy ieadeis, though age, gendei, and education aie discussed
at vaiious points in the analyses. Te majoiity of the data stemfiomlaige, nationally
iepiesentative studies completed aftei the 2004 publication of the NEAs Reading at
Risk iepoit. Unless a footnote is piovided, souices foi all data in this Executive Sum-
maiy aie given with each accompanying chait oi table. All adult ieading scoies and
piociency iates iefei to the Depaitment of Educations piose liteiacy categoiy.
Caution should be used in compaiing iesults fiomthe seveial studies cited in this
publication, as the studies use dieient methodologies, suivey populations, iesponse
iates, and standaid eiiois associated with the estimates, and the studies often weie
designed to seive dieient ieseaich aims. No denite causal ielationship can be made
between voluntaiy ieading and ieading piociency, oi between voluntaiy ieading,
ieading piociency, and the ieadei chaiacteiistics noted in the iepoit. Finally, except
wheie book ieading oi liteiaiy ieading iates aie specically mentioned, all iefeiences
to voluntaiy ieading aie intended to covei all types of ieading mateiials.
Oce of Reseaich & Analysis
National Endowment foi the Aits
I:voouc:io
T
he National Endowment foi the Aits landmaik iepoit, Reading at Risk: A
Survey of Literary Reading in America, giew out of a peiiodic suivey the
agency has been conducting since 1982. Co-developed with an expeit teamof
statisticians, sociologists, and economists, the suivey collected data fiommoie than
17,000 adults, iepoited on eveiy demogiaphic gioup in the countiy, and ielied on
the U.S. Census Buieau foi its administiation. Key ndings of the iepoit occasioned
consideiable commentaiy fiom policy-makeis, educatois, libiaiians, jouinalists,
scholais, and aits and cultuial oiganizations. Tose statistics included:
Only 47 of adults iead a woik of liteiatuie (dened as a novel, shoit stoiy, play
oi poem) within the past yeai.
Tat guie iepiesented a 7-point decline in the peicentage of liteiaiy ieadeis
ovei a 10-yeai peiiod.
Liteiaiy ieading declined in both gendeis, acioss all education levels, and in
viitually all age gioups.
Te declines weie steepest in young adults, acceleiating at a gieatei iate than in
the geneial population.
Ameiicans weie not only ieading liteiatuie at a ieduced iatethey weie iead-
ing fewei books geneially.
Te NEAhad begun to outline the nations ieading habits as eaily as 1989, with the
publication of Reseaich Repoit #22, Who Reads Literature? Based on the 1985 Suivey
of Public Paiticipation in the Aits, that iepoit found 56of Ameiicans ieading litei-
atuie, a peicentage that diopped 10 points by 2002. But Reading at Risk was qualita-
tively dieient fiom piioi NEA iepoits foi the public concein it iaised. With its
distiessing snapshot of liteiaiy ieading in Ameiica, the study piovoked a national
conveisation in news media and classiooms, and among local, state, and fedeial agen-
cies, on how to iespond to the pioblem.
Te Aits Endowment iesponded by paitneiing with anothei fedeial agency, the
Institute of Museum and Libiaiy Seivices, and the iegional aits oiganization Aits
Midwest, to initiate the Big Read, a giassioots liteiaiy piogiam of unpiecedented
scale, designed to biing togethei Ameiican communities in the ieading and celebia-
tion of gieat imaginative liteiatuie. By the end of 2007, neaily 200 towns and cities
will have adopted the Big Read, and a national evaluation will identify the extent to
which the piogiam is helping to impiove the ieading iates of paiticipants.
Foi a moie extensive suivey of Ameiican ieading tiends since Reading at Risk, the
Aits Endowment will collect newdata on ieading as pait of the agencys 2008 Suivey
of Public Paiticipation in the Aits. Although ndings fiomthat study will not appeai
until late 2008, the agency has wanted to addiess seveial uigent questions iaised by
Reading at Riskat least those questions foi which data are available. Foi example,
how does leisuie ieading faie in the lives of young Ameiicans? (Only Ameiicans 18
yeais of age oi oldei weie included in the Reading at Risk study.) Also, what is the
ielationship between ieading foi pleasuie and ieading piociency? Finally, why does
ieading mattei to oui nation and communities?
Tis iepoit uses laige-scale, nationally iepiesentative data fiom a vaiiety of
souicesfedeial, academic, nonpiot, and commeicialto diawa pictuie of volun-
taiy ieading iates and liteiacy in Ameiican life. Te vast majoiity of these ndings
aiose only aftei the 2004 publication of Reading at Risk, and nevei have been col-
lected in a single naiiative about ieading in oui cultuie and oui times.
Te iepoit consists of thiee paits, subdivided into a total of 9 chapteis. Each chap-
tei begins with its conclusion: a statement whose tiuth is boine by subsequent data.
Each chaptei also includes a backgiound section pioviding context foi the analysis
and listing all souices.
Of all the vaiiables that can be examined in conjunction with leisuie ieading and
ieading skills, this iepoit is conceined chiey with age and education level. Te
potential ioles of iace, ethnicity, income level, oi leaining-ielated and othei disabil-
ities have not been consideied, in pait because iepoiting of those vaiiables is incon-
sistent fiom study to study. Also absent is a discussion of U.S. public libiaiies and
theii pait in piomoting ieading of all kinds. Te lack of ieliable national data on
libiaiy ciiculation iates foi ieading mateiialsas sepaiate fiomCDs and videotapes,
foi examplehas infoimed this decision.
Foi the most pait, footnotes have been avoided foi souices alieady cited in a chap-
teis backgiound section. Unless book-ieading is specically mentioned, study
iesults on voluntaiy ieading should be taken as iefeiencing all vaiieties of leisuie
ieading (e.g., magazines, newspapeis, online ieading), and not books alone. Unlike
Reading at Risk, this iepoit is not limited to the ieading piimaiily of liteiatuie, except
wheie that eailiei iepoit is cited.
An all-impoitant woid of caution: None of the data on reading prociency and
the personal or social attributes of readers should be regarded as drawing a
causal relationship between voluntary reading, reading skills, and other vari-
ables. As we note elsewheie, the longitudinal study oi iandomized, contiolled tiial
that would dene those ielationships is conspicuously absent. Yet the iecuiiing asso-
ciations between voluntaiy ieading and advanced ieading skills and othei benets
aie compelling in theii own iight.
Te Suivey of Public Paiticipation in the Aits, with its up-to-date iepoiting of
adult ieading iates, will commence in May 2008. By then, many of the teenage gioups
discussed in this iepoit, those showing the gieatest declines in voluntaiy ieading and
in piociency, will have aged into the population eligible foi the suivey. One hopes
theii engagement with books and othei ieading mateiials will have impioved by that
point. Until then, like stock analysts who must watch and iecoid eveiy uctuation in
the maiket, though they cannot piedict outcomes with ceitainty, we oei this iepoit
captuiing some of the most cuiient and ieliable statistical infoimation on ieading.
Sunil Iyengai
Diiectoi, Reseaich & Analysis
National Endowment foi the Aits
Fotostudio FM/zefa/Corbis
CHAPTER ONE
1. Young adults are reading fewer books in general.
2. Reading is declining as an activity among teenagers.
BACKGROUND
W
hen the National Endowment foi the Aits ieleased its Reading at Risk
iepoit in 2004, one of the moie widely discussed ndings was that
declines in liteiaiy ieading weie shaipest in young Ameiican adults.
Twenty yeais eailiei, 18- to 34-yeai-olds had been the age gioup most likely to iead
liteiatuie. Te 2002 Suivey of Public Paiticipation in the Aits ievealed they weie now
the least likely (with the exception of Ameiicans aged 65 and oldei, a gioup whose
aveiage ieading iates may have been stied by health issues).
Each yeai of the suivey since 1992, the Aits Endowment has asked not only about
paiticipation in liteiaiy ieading, but also about the ieading of books in geneial.
Among young adults in 2002, these data show tiends similai to those foi liteiaiy
ieading iates. Te book-ieading iate of 18- to 24-yeai-olds was only 52, signicantly
belowthe peicentage of all Ameiicans who iead a book in 2002 (57). Heie, as with
liteiaiy ieading, 18- to 24-yeai-olds showed one of the steepest peicentage point
declines foi all age gioupsa 7-point diop fiom1992 to 2002, iepiesenting a loss of
2.1 million potential ieadeis.
By contiast, the gioup with the gieatest peicentage of book ieadeis within it (61)
was 4554 yeais old. Peihaps not suipiisingly, this age gioup was the one most likely
to iead novels, shoit stoiies, poetiy, and plays. Indeed, a cential nding of Reading at
Risk was that liteiaiy ieading iates declined substantially foi all adult age gioups
Src:io Or
Youth Voluntaiy Reading Patteins
Table 1A. Percentage of Adults Who Read a Book Not Required for Work or School,
by Age Group
Age 1992 2002 Change
1824 59% 52% -7 pp
2534 64% 59% -5 pp
3544 66% 59% -7 pp
4554 64% 61% *-3 pp
5564 59% 58% *-1 pp
6574 55% 54% *-1 pp
75+ 42% 44% *+2 pp
* no statistically significant change from 1992
Source: National Endowment for the Arts, Survey of Public Participation in the Arts
undei 45.
1
Tis phenomenon extended to book ieading of all types. See Table 1A.
Given this pattein of diminished ieading in the U.S.and the concentiation of the
tiend in 18- to 44-yeai-olds liteiaiy and book-ieading iatesit is ieasonable to ask
whethei such declines also appeai in childien and teenageis. Because the Suivey of
Public Paiticipation in the Aits (SPPA) is limited to U.S. adults, it cannot answei this
question. Foitunately, othei laige-scale studies have investigated ieading habits in
Ameiicans undei 18. Tis chaptei consideis data fiom thiee piimaiy souices:
U.S. Depaitment of Education: NAEP 2004 Trends in Academic Progress: ree
Decades of Student Performance in Reading and Mathematics, 2005.
Te Heniy J. Kaisei Family Foundation: Generation M: Media in the Lives of
8-18 Year-Olds, (#7251), 2005.
Highei Education Reseaich Institute, Univeisity of Califoinia, Los Angeles:
American Freshmen: Forty-Year Trends 19662006, 2007.
Like most of the data in this iepoit, the iesults of all thiee studies became available
aftei Reading at Risk was published. Although none identify the iates specically foi
liteiaiy ieading among childien and adolescents, the studies do iepoit peicentages of
voluntaiy (oi leisuie) ieadeis among this population.
What Education Research Tells Us About Leisure Reading
Te Depaitment of Educations NAEP, known as the Nations Repoit Caid, has
tiacked achievement test scoies of elementaiy, middle, and high school students foi
a iange of subjects since 1969. In addition to this main assessment, conducted
nationally and at the state level, the NAEP piovides a long-teimtiend assessment of
9-, 13-, and 17-yeai-olds in ieading and mathematics. Te tiend assessment is based
on a nationally iepiesentative sample and is conducted ioughly eveiy foui yeais. In
2004, the most iecent yeai of the tiend assessment, 38,000 students paiticipated in
the ieading component.
Adiscussion of ieading achievement tiends, as chaited by NAEP, follows in Chap-
tei Five of this iepoit. Foi oui immediate puipose, howevei, the NAEP long-teim
assessment oeis valuable infoimation about childhood and teenage ieading iates.
Tese data aie made available because apait fiomtesting students piogiess in iead-
ing, the assessment asked students to iepoit contextual vaiiables such as time spent
on homewoik, the numbei of pages iead foi school and homewoik, and the amount
of time spent ieading foi fun.
2
Table 1B shows the fiequency of leisuie ieading by 9-, 13-, and 17-yeai-olds in
thiee yeais: 1984, 1999, and 2004. Students on the lowest end of the age scale evinced
not only the highest leisuie ieading iate foi all thiee yeaismoie than half of 9-yeai-
olds iepoited ieading almost eveiy day, compaied with 2835 of 13-yeai-olds
and 2231of 17-yeai-oldsbut one that did not altei signicantly with each test
peiiod. By 2004, 9-yeai-olds had the smallest peicentage of non-ieadeis (students in
the nevei oi haidly evei iead categoiy), and foi all thiee yeais they had the smallest
peicentage of infiequent ieadeis (a few times a yeai) ielative to the two othei age
gioups.
Chaptei Seven desciibes how the consistently highei leisuie ieading iates in 9-
yeai-olds coiiespond with impiovements in theii ieading achievement scoies. Foi
now, 13- and 17-yeai-old leisuie ieading iates invite a closei compaiison with those
of the youngei age gioup.
1
NEA, Reading at Risk: A Survey
of Literary Reading in America,
2004, 2628.
2
Nine-, 13- and 17-yeai-olds weie
asked How often do youiead
foi fun on youi own time?
Among 19842004 tiend data foi 9-yeai-old ieading iates, the one statistically sig-
nicant nding was a giowth in infiequent ieadeis by two peicentage points (fiom3
in 1984 to 5in 2004). Foi Ameiicans at the eaily and late stages of adolescence, how-
evei, statistically signicant changes appeaied at each extieme of the fiequency band.
Te peicentage of avid ieadeis (almost eveiy day) slumped foi 13- and 17-yeai-olds,
fiom35to 30and fiom31to 22, iespectively. Conveisely, both age gioups expe-
iienced a jump in the peicentage of non-ieadeis, fiom8to 13of 13-yeai-olds and
fiom 9 to 19 of 17-yeai-olds. Te lattei giowth iate iepiesents a doubling of the
percentage of 17-year-olds who never or hardly ever read for fun.
Te changes in leisuie ieading iates aie disconceiting. In the two adolescent age
gioups, the iates have declined to such an extent that by 2004 less than one-thiid of
13-yeai-olds iead foi fun almost eveiy day. Te coiiesponding guie foi 17-yeai-
olds, combined with a giowing peicentage of non-ieadeis in both teen gioups, leads
us to ask: What factois account foi the ielatively stable leisuie ieading iates among
9-yeai-olds?
Moie to the point, why aie voluntaiy ieading iates consistently highei in 9-yeai-
olds than in 13- and 17-yeai-olds? It may be unwise to discount biological and social
developmental factois in a childs tiansition to adolescence: the teens giowing need
foi peei appioval, his oi hei bid foi gieatei independence, the likelihood of less
paiental supeivision, pubeity itself. Still, none of these factois necessaiily conict
with a voluntaiy ieading lifestyle. Noi aie they sucient to explain the extent of the
dieience in ieading iates.
If the low ieading iates foi 13- and 17-yeai-olds weie constant ovei time, one
might be tempted to call them chaiacteiistic of those age gioups. Yet not only have
teenagei ieading iates iemained well below those of 9-yeai-olds, teen ieading iates
have diminished in the past two decades.
Table 1B. Percentage of Students Ages 9, 13, and 17, by Frequency of Reading for Fun: 1984, 1999, and 2004
Age 9
Almost Once or twice Once or twice A few times Never or
every day a week a month a year hardly ever
1984 53% 28% 7% 3% 9%
1999 54% 26% 6% 4% 10%
2004 54% 26% 7% 5% 8%
Age 13
1984 35% 35% 14% 7% 8%
1999 28% 36% 17% 10% 9%
2004 30% 34% 15% 9% 13%
Age 17
1984 31% 33% 17% 10% 9%
1999 25% 28% 19% 12% 16%
2004 22% 30% 15% 14% 19%
Shading of a column denotes statistically significant changes between 1984 and 2004.
Voluntary Versus Compulsory Reading
Teie is anothei fiequently voiced explanation foi declines in ieading fiomchildhood
to adolescence. It is sometimes aigued that as childien entei and piogiess thiough
high school, they toil undei moie iigoious couisewoik, which, along with extiacui-
iiculai activities, tend to sap time fiom leisuie ieading.
3
(As shown latei in this
iepoit, a similai view must be faced when chaiacteiizing ieading tiends in college
students.)
Te NAEP long-teimtiend assessment is a handy iesouice to consult when inves-
tigating this claim. Table 1Cshows the numbei of pages iead daily by 9-, 13-, and 17-
yeai-oldsin school oi foi homewoikfoi the yeais 1984, 1999, and 2004. Te
peicentages suggest that compulsoiy ieading has incieased foi both 9- and 13-yeai-
olds. Fiom 1984 to 2004, a statistically signicant diop in the peicentage of 9-yeai-
olds ieading 010 pages a day is accompanied by a piopoitionate inciease in the
peicentage ieading 16 oi moie pages a day.
Tiiteen-yeai-olds also expeiienced a 20-yeai decline in the peicentage of students
ieading 010 pages a day foi school, with coiiesponding incieases in the 1620 and
moie than 20 pages-daily categoiies. Tis lattei peicentage is neaily twice as gieat
as the 1984 level.
By compaiing Table 1Cwith 1B, we conclude that a giowth in school-ielated iead-
ing foi 9-yeai-olds does not appeai to hindei theii voluntaiy ieading iate, but that
foi 13-yeai-olds, the ielationship between compulsoiy and voluntaiy ieading is alto-
gethei dieient. Tiiteen-yeai-olds iead moie foi school than in pievious yeais, and
they also iead less foi pleasuie.
Yet this conict is not appaient in oldei teens. In 17-yeai-oldsthe gioup whose
voluntaiy ieading iates faied the woistthe percentage of students at all ve levels
of compulsory reading has remained largely constant foi the thiee test peiiods.
3
Te Heniy J. Kaisei Family
Foundation, Generation M: Media
in the Lives of 818-Year-Olds
(#7251), 2005, 26.
Table 1C. Percentage of Students Ages 9, 13, and 17, by Pages Read Per Day in School and for Homework: 1984,
1999, and 2004
Age 9
5 or fewer 6 to 10 11 to 15 16 to 20 More than 20
1984 36% 25% 14% 13% 13%
1999 28% 24% 15% 14% 19%
2004 25% 21% 13% 15% 25%
Age 13
1984 27% 34% 18% 11% 11%
1999 23% 31% 18% 13% 16%
2004 21% 26% 18% 14% 21%
Age 17
1984 21% 26% 18% 14% 21%
1999 23% 24% 17% 14% 22%
2004 21% 24% 17% 15% 23%
Shading of a column denotes statistically significant changes between 1984 and 2004.
4
Ibid. Howevei, the peicentage of
11- to 14-yeai-olds who iead a
book foi at least 30 minutes the
pievious day does not diei signif-
icantly fiom the peicentage of 15-
to 18-yeai-olds who did, as that
sentence would appeai to indicate.
5
Tis table does not display the
iepoited time spent ieading othei
mateiials, including magazines
and newspapeis. Foi both of these
media, the peicentage of ieadeis
among the 11- to 14-yeai-old
and 15- to 18-yeai-old gioups
is greater than foi the 8- to 10-
yeai-old gioupbut only in the
5-minutes-oi-moie categoiy.
Foity-seven peicent of 15- to 18-
yeai-olds iead a magazine foi at
least 5 minutes, compaied with
35 and 54 of the 810 and
1114 age gioups, iespectively.
Foity-thiee peicent of 15- to 18-
yeai-olds iead a newspapei foi at
least 5 minutes, compaied with
21 and 35 of the 810 and
1114 age gioups, iespectively.
Foi both media, theie was no
statistically signicant dieience
in the peicentage of 11- to 14-
yeai-old and 15- to 18-yeai-old
childien who iead foi at least 5
minutes. Indeed, the peicentage of
8- to 10-yeai-olds who iead any of
the thiee mediabooks, maga-
zines, oi newspapeisfoi at least
5 minutes (73) did not diei in
compaiison with the 11- to 14-
and 15- to 18-yeai-old gioups.
Regaiding newspapei and
magazine ieading foi at least 30
minutes the pievious day, theie
weie no statistically signicant
dierences in the peicentages
of any of the age gioups. Tus,
foi 8- to 10-yeai-olds, the peicent-
age who iead any of the thiee
mediabooks, magazines, oi
newspapeisfoi at least 30 min-
utes (51) was ioughly the same
as foi the two oldei age gioups.
6
Fiom tiend data piovided by
UCLAs Highei Education
Reseaich Institute.
Indeed, by 2004, the peicentage of 17-yeai-olds in the moie than 20 pages-a-day
categoiy has failed to suipass that of 9-yeai-olds.
Tis obseivation does not pieclude the possibility that expanded school-ielated
activities among teens aie displacing time spent on leisuie activities such as ieading
but it does complicate that viewconsideiably. In this study, 17-yeai-olds aie the only
age gioup that a) lacked giowth in the peicentage ieading moie than 15 pages daily
foi school oi homewoik and b) sawno attiition of the peicentage ieading fewei than
11 pages daily foi school oi homewoik. At the same time, they weie the only age gioup
to expeiience a doubling of the peicentage that nevei oi haidly evei ieads foi pleasuie.
Without attempting to quantify ieading foi school oi homewoik, anothei study
shows a piogiessively lowei iate of book ieading foi pleasuie acioss age gioups fiom
8 to 18 yeais old.
Generation M: Media in the Lives of 818 Year-Olds iesulted fioma nationally iep-
iesentative suivey of 2,032 thiid- thiough twelfth-giade students, supplemented by
694 seven-day media-use diaiies. Although the study aimed piimaiily to examine
students access to a host of tiaditional and electionic media, and theii ielated use
patteins, it also shed light on the ieading habits of teenageis.
Generation Miepoited that less than half of 8- to 18-yeai-olds spent at least 5 min-
utes of the pievious day on the leisuie ieading of books. Yet within that gioup, 63
of 8- to 10-yeai-olds iead books foi 5 minutes oi moie, while only 44 of 11- to 14-
yeai-olds did so. At the latei stages of adolescence, that peicentage dwindled to 34.
Indeed, with each successive inciease in age, both the piopoition of kids who engage
in leisuie book ieading and the piopoition who iead books foi at least 30 minutes
decieases signicantly, the study authois conclude befoie tentatively asciibing those
disciepancies to incieased ieading foi school.
4
See Table 1D.
5
If heaviei couise iequiiements weie eioding the leisuie ieading time of high school
students, one might expect the factoi to be documented by laige population suiveys
of teen expeiiences at home and school.
As it happens, a long-teimtiend analysis of high school seniois found that in 2006
only 33 of them iepoited spending 6 oi moie houis a week on homewoik, com-
paied with 47in 1987.
6
Te same analysis ievealed, howevei, that high school sen-
iors leisure reading rates have failed to improve in the last 13 years foi which
the suivey has tiacked this activity. In 1994, less than 20 of seniois iepoited they
spent no time ieading foi pleasuie on a typical week. In 1997, that guie jumped to
25 and has lingeied theie evei since.
Meanwhile, the peicentage ieading 6 oi moie houis a week foi pleasuie has
diopped by two points, iepiesenting a 17 iate of decline. Table 1E shows the pei-
centage of high school seniois engaged in leisuie ieading in 1994 and 2006 foi the
time amounts measuied by the suivey. Foi each yeai of the suivey, students who iead
02 houis a week made up a majoiity.
Table 1D. Percentage Who Read Books for Fun the Previous Day: 2004
8- to 18- 8- to 10- 11- to 14- 15- to 18-
year-olds year-olds year-olds year-olds
Read for at least 5 minutes 46% 63% 44% 34%
Read for at least 30 minutes 30% 40% 27% *26%
* No statistically significant change from the percentage directly to the left
Source: The Henry J. Kaiser Family Foundation, Generation M: Media in the Lives of 8-18 Year Olds, (# 7251), 2005
Chait 1F tiacks yeai-ovei-yeai changes in the peicentage of students who iead 6 oi
moie houis a week foi pleasuie. Foi the sake of compaiison, a line is also given foi
the peicentage of high school seniois who did homewoik foi 6 oi moie houis a week.
As illustiated by two veitical axes, both activities display a downwaid tiend fiom
1994 thiough 2001.
Compaiisons such as these do not solve the puzzle of why leisuie ieading iates aie
declining among teenageis, yet they do suggest that compulsoiy ieading tiends lack
a cleai and obvious ielationship with voluntaiy ieading patteins. As Chaptei Two
explains, ielatively low ieading iates peisist thioughout the teens entiy into college
and adulthood.
Table 1E. Percentage of High School Seniors Who Read for Pleasure, by Hours
Per Week
1994 2006 Change Rate of change
None 20% 25% +5 pp +25%
Less than one hour 25% 25% 0 pp 0
1 to 2 hours 26% 25% -1 pp -4%
3 to 5 hours 17% 16% -1 pp -6%
6 or more 12% 10% -2 pp -17%
Numbers do not total 100% due to rounding.
2.0%
3.0%
4.0%
5.0%
1994 1996 1998 2000 2002 2004 2006
45%
40%
35%
30%
25%
%
R
e
a
d
i
n
g
f
o
r
P
l
e
a
s
u
r
e
6
o
r
M
o
r
e
H
o
u
r
s
a
W
e
e
k
%
W
h
o
D
i
d
H
o
m
e
w
o
r
k
6
o
r
M
o
r
e
H
o
u
r
s
a
W
e
e
k
5%
42%
4%
33%
Chart 1F. Percentage of High School Seniors Who Read for Pleasure and/or
Did Homework 6 or More Hours Per Week: 19942006
7
U.S. Census Buieaus 2004 Cui-
ient Population Suivey, Depait-
ment of Educations 2006 Digest of
Education Statistics (National
Postsecondaiy Aid Study), Table
180.
CHAPTER TWO
3. College attendance no longer guarantees active reading habits.
BACKGROUND
I
n 2005, appioximately 8.5 million students weie eniolled as undeigiaduates in
U.S.-based foui-yeai colleges and univeisities. By paitaking of highei education,
those individuals weie on tiack to join the nations 52 million college giaduates,
who account foi 28 of Ameiicans ages 25 yeais oi oldei.
7
College attendance, although fai fiom univeisal, is often depicted as a iite of pas-
sage into independent adulthood. (Te woid Commencement, and the ceiemonies
it desciibes, maiks the beginning of that tiansition.) Tioughout the histoiies of
modein univeisities, college leaining has been poitiayed as a seiies of classioomand
out-of-classioomexpeiiences and iesponses that equip the student foi paiticipation
in adult life.
In Te Idea of a Univeisity (1854), John Heniy (Caidinal) Newman distinguishes
between didactic leaining and the behavioial fiamewoik that must be cultivated if
education and not meie extiinsic oi accidental advantage is to pievail:
[Knowledge] is an acquiied illumination, it is a habit, a peisonal possession, and
an inwaid endowment. And this is the ieason why it is moie coiiect, as well as
moie usual, to speak of a Univeisity as a place of education, than of instiuction,
though, when knowledge is conceined, instiuction would at ist sight have
seemed the moie appiopiiate woid.But education is a highei woid, it implies
an action upon oui mental natuie, and the foimation of chaiactei, it is some-
thing individual and peimanent.
Newmans distinction between instiuction and education pioves helpful when
we considei voluntaiy ieading as a stiategy to ieinfoice the skills and habits needed
foi lifelong leaining. Tiee laige population studies gauge the intensity of that
engagementof ieading beyond the cuiiiculumduiing this foimative peiiod. Tey
aie:
Highei Education Reseaich Institute, Univeisity of Califoinia, Los Angeles:
Findings from the 2005 Administration of Your First College Year (YFCY) Sur-
vey: National Aggregates, 2007.
Highei Education Reseaich Institute, UCLA: Findings from the 2005 College
Senior Survey (CSS): National Aggregates, 2007.
Indiana Univeisity Bloomington: National Survey of Student Engagement 2006
Institutional Report, 2006.
UCLA Freshman and Senior Surveys
Te ist two iepoits stem fiom annual suiveys conducted by UCLAs Highei Edu-
cation Reseaich Institute (HERI), which also pioduced the tiend analysis discussed
in Chaptei One (see pp. 3132). As pait of the Coopeiative Institutional Reseaich
Piogiam, a longitudinal study billed as the nations oldest and laigest empiiical study
8
See HERI website (http:iiwww.
gseis.ucla.eduiheiiiheii.html). In
2006, the baseline suivey obtained
iesponses fiom 271,441 ist-yeai
college students at 393 U.S. col-
leges and univeisities.
9
HERI, Findings from the 2005
Your First College Year (YFCY)
Survey: National Aggregates, 2007,
20. Othei notable declines aiose
in students iepoited time spent
playing spoits, exeicising, paitici-
pating in student clubs oi gioups,
oi watching television.
of highei education, UCLA collects baseline data each yeai fiom incoming college
students.
8
Te baseline data can be analyzed sepaiately oi in tandem with two othei annual
suiveys: UCLAs Your First College Year suivey and the College Senior Survey, which
poll college students appioaching the end of theii fieshman and senioi yeais, iespec-
tively. Because the thiee suiveys yield a wealth of statistics on student life expeiiences
fiom twelfth giade to collegeand fiom the undeigiaduates ist thiough nal
yeaithey allow us to assess ieading habits at key points along the continuum.
In the annual iepoit foi the 2005 Your First College Year suivey, ieseaicheis iden-
tied ieading foi pleasuie as one of seveial leisuie activities that have sustained
notable declines as high school students entei college and piogiess thioughout
theii ist yeai.
9
In the fall of 2004, 27of incoming college fieshmen iepoited having
iead foi pleasuie foi thiee oi moie houis pei week duiing theii senioi yeai in high
school. By the spiing of 2005when those students weie set to nish theii ist yeai
of collegeonly 15 iepoited that level of ieading intensity. See Table 2A.
Te situation does not impiove as students advance thiough college. Although the
peicentage of college seniois in 2005 who did no ieading foi pleasuie was 4 points
less than that of college fieshmen in the same yeai (35 veisus 39), a longitudinal
peispective piompts a moie sobeiing view.
Table 2B shows that the college seniois of 2005 had a gieatei likelihood of ieading
less on any given week than when they weie high school seniois. Twenty-one peicent
of incoming college fieshmen iepoited ieading nothing foi pleasuie duiing theii sen-
ioi yeai in high school, which, foi the majoiity of students, was pie-2002. By the time
that population ieached senioi yeai in college, the percentage of nonreaders had
climbed by 14 points, to 35%. Moieovei, only 14of 2005 college seniois iepoited
ieading weekly foi 3 houis oi moie. Tat guie maiks a 10-point diop fiomthe pei-
centage who iead foi the same amount of time as high school seniois.
Table 2A. First-Year College Students: Time Spent Reading for Pleasure
Reading Percentage as high school Percentage as college
per week: seniors in 2004 freshmen in 2005 Change
None 21% 39% +18 pp
Less than 1 hour 26% 26% 0 pp
1 to 2 hours 27% 20% -7 pp
3 or more hours 27% 15% -12 pp
Table 2B. College Seniors: Time Spent Reading for Pleasure
Reading Percentage as high school Percentage as college
per week: seniors seniors in 2005 Change
None 21% 35% +14 pp
Less than 1 hour 28% 28% 0 pp
1 to 2 hours 27% 23% -4 pp
3 or more hours 24% 14% -10 pp
10
See NSSE Web site: Oui Oii-
gins and Potential (http:iinsse.
iub.eduihtmlioiigins.cfm).
11
Te coiiesponding item on the
questionnaiie is Numbei of
books iead on youi own (not
assigned) foi peisonal enjoyment
oi academic eniichment duiing
the cuiient school yeai.
National Survey of Student Engagement
Results fiom anothei laige-scale, institutional suivey tell a dieient stoiy. Indiana
Univeisity has polled iandomsamples of ist-yeai and senioi college students eveiy
yeai since 1999, focusing its questions on items that aie known to be ielated to
impoitant college outcomes.
10
Unlike the UCLAsuiveys, the National Suivey of Stu-
dent Engagement (NSSE) asks about leisuie ieading in teims of volume of books
iead, not houis spent ieading.
NSSE is not designed as a longitudinal studyand theiefoie does not tiack the
same cohoit of students as they pass thiough collegebut the ieseaichei can com-
paie ist-yeai and senioi book-ieading iates foi a given yeai. Towaid the end of theii
fieshman yeai, 27 of college students iepoited ieading zeio books foi pleasuie oi
eniichment in 2007.
11
Foi outgoing college seniois, the coiiesponding guie is 21.
Similaily, while only 18 of ist-yeai college students had iead ve oi moie books
duiing the school yeai, college seniois iepoited doing so at a iate of 25. See Table
2C.
Despite the appaiently signicant gains in college senioi ieading iates when com-
paied with students ioughly foui yeais theii junioi, this analysis does not iepoit
whethei the same seniois iead fewei oi gieatei numbeis of books than in theii ist
yeai of college. Similaily, the study does not ieveal whethei college students expeii-
enced a change fiomtheii ieading levels in high school. Still, if we accept that volun-
taiy ieading habits aie cential to a libeial aits education, then suiely it is tioubling
to nd that the majoiity of fieshmen and seniois alike iead only 14 books foi pleas-
uie thioughout the entiie school yeai oi they iead no unassigned books at all.
As when ieviewing ieading iates of high school students, we might ask whethei
inciementally moie challenging couise iequiiements aie stunting the giowth of vol-
untaiy ieading in undeigiaduates. Te question cannot be answeied with the NSSE
data alone. Still, the 2007 data showthat seniois aie not assigned moie college texts,
on aveiage, than fieshmen of the same yeaisee Table 2Dand that seniois, on
aveiage, do not spend moie time on class piepaiation. Table 2E claiies the lattei
point.
Fiom these data, it still is plausible that college seniois, in aggiegate, aie assigned
moie iigoious couisewoik in geneial than ist-yeai students, though not ieected in
the amount of assigned texts. Yet this is by no means the only way to inteipiet the
data. Te NSSE iesults do not explain, aftei all, whethei the students completely iead
eveiy assigned text, whethei the texts ianged in complexity fiomfieshman to senioi
yeai, oi whethei the piesciibed ieading mateiials giew piogiessively moie complex
duiing the same college yeai.
Similaily, we lack a numbei foi othei, non-ieading assignments and activities. At
Table 2C. Comparison of First-Year and Senior College Student Reading Rates in 2007
Percentage of Percentage of Gap between
first-year students seniors groups
Number of unassigned None 27% 21% -6 pp
books read 14 55% 54% -1 pp
5 or more 18% 25% +7 pp
Source: Indiana University Bloomington, National Survey of Student Engagement
12
See Table 2C, p. 35.
the end of this exeicise, we aie left with a cheeiless statistic: 74%80% of college
freshmen and seniors read 04 books on their own duiing the school yeai.
12
We aie also left with a fuithei question: if we cannot ciedit school and homewoik
assignments entiiely with displacing eaily ieading habits, then what othei factois aie
iesponsible? To exploie these issues, we must tuin to a fedeial study of how Ameii-
cans spend theii leisuie time.
Table 2E. Comparison of Time Spent on College Preparation in 2007
Preparing for class 0 hours/wk 0% 0% 0 pp
(studying, reading, 15 hours/wk 17% 18% +1 pp
writing, doing 610 hours/wk 27% 26% -1 pp
homework, etc.) 1115 hours/wk 22% 19% -3 pp
16 hours or more/wk 34% 35% +1 pp
Table 2D. Comparison of College-Prescribed Reading Levels in 2007
Number of assigned None 1% 1% 0 pp
textbooks, books, or Between 14 22% 28% +6 pp
book-length packs of Between 510 44% 39% -5 pp
course readings Between 1120 24% 20% -4 pp
More than 20 10% 12% +2 pp
13
Robinson, John P. and Godbey,
Geoiey, Time for Life: e Sur-
prising Ways Americans Use eir
Time: Penn State Piess, Univeisity
Paik, PA. 1997, 67.
14
Ibid.
CHAPTER THREE
4. Teens and young adults spend less time reading than people of other age
groups.
5. Even when reading does occur, it competes with other media.
BACKGROUND
T
he iapid technological shifts that maiked the second half of the last centuiy
bieakthioughs in electionics, computeis, and communicationsieveibeiated
fai beyond the elds wheie they occuiied. Paiallel developments in medicine,
commeice, media, and manufactuiing have continued to benet the aveiage con-
sumei. Yet a subtlei ievolution is dictating the way Ameiicans live and woik today.
With the advent of time-saving technologies, we have moie time to spend on
leisuiebut we also have moie ways to spend time on woik. (Blackbeiiies, pageis, and
cell-phones attest to this paiadox, as does the giowing populaiity of telecommuting,
which collapses conventional boundaiies of home and woik.) To some extent, the Dig-
ital Age meiely acceleiated an eailiei piocess. Moie fundamental changes in woik and
leisuie patteins had oiiginated with 18th- and 19th-centuiy industiialization, which
heightened public awaieness of time as a commodity to be managed.
In eaily-to-mid-20th-centuiy Ameiica, two tiends placed time management undei
gieatei sciutiny than befoie. Fiist weie impiovements to manufactuiing eciencies,
iepiesented by the automobile industiy and the assembly line. Innovations in scalable
pioduction, involving the subdivision of time and tasks into disciete units, ensuied
Ameiicas competitiveness in a global maiket.
Te second change was the commeicialization of leisuie in the two decades aftei
Woild Wai II, a peiiod of uniivalled piospeiity foi the nation. Pioduct vendois and
adveitiseis began vying moie aggiessively foi the leisuie houis of a giowing middle-
class. Tey also taigeted the fiee time of Ameiicas laigest geneiational cohoit evei,
the Baby Boomeis.
Roughly concuiient with those tiends, academic ieseaicheis, coipoiations, and
public policy planneis began to ciave a moie sophisticated undeistanding of how
Ameiicans use leisuie time. In 1954, moie than 8,000 Ameiicans paiticipated in a
time-use study conducted by the Mutual Bioadcasting Coipoiation.
13
Tis exeicise
involved time diaiies in which subjects iepoited all theii activities within a two-
day peiiod. Time diaiies weie again used in a 1965 multinational study including
moie than 1,200 Ameiicans. Subsequent majoi time-use studies occuiied in 1975,
1985, and thioughout the 1990s.
14
Tis chaptei examines leisuie ieading habits in the context of moie iecent time-
use studies. Te data deiive fiom foui independent souicesa piivate foundation,
two univeisity ieseaich teams, and the U.S. goveinment:
Depaitment of Labois Buieau of Laboi Statistics: American Time Use Survey
(ATUS), 2006.
Univeisity of MichiganiAnn Aibois Institute foi Social Reseaich: Changing
Times of American Youth: 19812003, 2004.
15
Foi details, see American Time
Use Survey Oveiview section of
BLS website (http:iiwww.bls.govi
tusioveiview.htm#puipose).
16
Socializing and communicat-
ing involves 1 houi and 7 minutes
of Ameiicans aveiage leisuie
timebut only on weekends and
holidays. On weekdays, the activ-
ity consumes 36 minutes.
Indiana Univeisity Bloomington: 2006 High School Survey of Student
Engagement (HSSSE), 2007.
Te Heniy J. Kaisei Family Foundation: Media Multitasking Among American
Youth: Prevalence, Predictors and Pairings, (# 7592), 2006.
Federal Data on Time Use Patterns
Since 2003, the Depaitment of Labois Buieau of Laboi Statistics (BLS) has spon-
soied the annual American Time Use Survey, conducted by the U.S. Census Buieau.
Consisting of telephone inteiviews, this nationally iepiesentative suivey included
about 13,000 people in 2006. Te inteiviewees, who must be at least 15 yeais old, aie
asked about the activities they did duiing the last 24 houis. (To avoid obtaining vaiied
iesponse iates thioughout the week, the inteiviews aie pie-assigned.)
15
Te suivey iesponses aie then used to iepoit the aveiage numbei of houis and
minutes pei day spentby the total civilian population oi a demogiaphic gioupin
peifoiming vaiious activities. BLS designates seveial categoiies of majoi activity,
including leisuie and spoits, of which ieading is a pait.
Accoiding to the suivey iesults foi 2006, Ameiicans leisuie and spoits activities
accounted foi an aveiage of 4 houis and 32 minutes on weekdays and 6 houis and 22
minutes pei day on weekends and holidays. Foi teenageis and veiy young adults, the
shaie of leisuie time is slightly gieatei, exceeding that foi all othei age gioups undei
55. See Table 3A.
Befoie we considei the aveiage numbei of houis pei day spent ieading, it is woith
knowing what othei types of activity compose the leisuie and spoits categoiy. Tey
aie: paiticipating in spoits, exeicise, and iecieation, socializing and communicat-
ing, watching TV, ielaxingithinking, playing games and computei use foi
leisuie, and othei leisuie and spoits activities, including tiavel.
When we ieviewthe aveiage numbei of houis that Ameiicans daily devote to these
activities, howevei, an anomaly aiises. As Table 3B ieveals, only one categoiy of
leisuie activity claims moie than an houi of Ameiicans daily timeclaims, indeed,
moie than 2 houis on weekdays and, on weekends, moie than 3. Tis activity is TV-
watching.
16
On aveiage, TV-watching consumes about half of the total daily leisure time
of all Americans ages 15 and older. Equally signicant, of all the leisuie categoiies,
only TV-watching is wholly dependent on electionic media. Foi example, the iecie-
ation in spoits, exeicise, and iecieation may iepiesent many dieient types of
activity. Likewise, playing games and computei use foi leisuie is not iestiicted to
computei games oi computei use. Socializing and communicating is desciibed as
face-to-face social communication and hosting oi attending social functions.
17
How does TV-watching time compaie with time spent ieading? Table 3C shows
two sets of columns: the amounts of time spent ieading and TV-watching by the vai-
ious age gioups. On weekdays, most Ameiicans undei 55 watch about 2 houis of TV.
(Oldei adults watch signicantly moie.) By contiast, this gioup spends less than 20
minutes pei weekday on ieading, with teens and young adults of 1534 yeais old
spending only 79 minutes.
Nowsee Table 3D, which gives the same time amounts as a peicentage of all leisuie
time. Although all age gioups iead fai less than they watch TV, we may take heait
that 15- to 24-yeai-olds spend a lowei peicentage of theii leisuie time, ielative to
othei age gioups, on TV-watching. Yet 15- to 24-year-olds still spend less than 3%
of their daily leisure time reading, and 25- to 34-yeai-olds spend ioughly 4.
17
BLS economic news ielease,
Ameiican Time Use Suivey
2006 Results, Technical Note.
Academic Time-Use Studies
In 1981, Univeisity of Michigan ieseaicheis began conducting nationally iepiesenta-
tive time-use studies of the behavioi patteins of childien and adolescents. Te data
involve Ameiican youth ages 617, and the studies allowcompaiisons of vaiious age
subgioups.
Table 3B. Hours and/or Minutes Per Day Spent on Leisure and Sports Activities,
by Activity
2006 Annual Averages
Ages 15 Years and Over Weekdays Weekends and holidays
Participation in sports, exercise and recreation 0:16 0:20
Socializing and communicating 0:36 1:07
Watching TV 2:21 3:06
Reading 0:20 0:26
Relaxing and thinking 0:17 0:20
Playing games and computer use for leisure 0:18 0:22
Other leisure and sports activities, including travel 0:24 0:41
Table 3C. Hours and/or Minutes Per Day Spent Watching TV or Reading
Hours/minutes spent watching TV Hours/minutes spent reading
Weekdays Weekends Weekdays Weekends
and holidays and holidays
Total, 15 years and over 2:21 3:06 :20 :26
15 to 24 years 1:57 2:33 :07 :10
25 to 34 years 1:55 2:51 :09 :11
35 to 44 years 1:53 2:39 :12 :16
45 to 54 years 2:07 3:02 :17 :24
55 to 64 years 2:35 3:35 :30 :39
65 years and over 3:56 4:10 :50 1:07
Table 3A. Hours and/or Minutes Per Day Spent on Leisure and Sports Activities,
by Age Group
Weekdays Weekends and holidays
Total, 15 years and over 4:32 6:22
15 to 24 years 4:39 6:33
25 to 34 years 3:38 5:52
35 to 44 years 3:34 5:20
45 to 54 years 3:54 5:59
55 to 64 years 4:47 6:54
65 years and over 7:08 7:57
18
F. Tomas Justei, et al., Univei-
sity of Michigans Institute foi
Social Reseaich, Changing Times
of American Youth: 19812003,
2004. Table 1.
Foi a 20022003 study, published in late 2004, ieseaicheis inteiviewed 2,017 fam-
ilies who had paiticipated in a 1997 time-use study. Te 20022003 inteiviews gen-
eiated data foi neaily 3,000 childien and adolescents. Co-sponsoied by the National
Institutes of Healths National Institute of Child Health &Human Development, the
study had childien and teens ll out 24-houi time-use diaiies foi iandomly selected
weekdays and weekends, with 6- to 9-yeai-olds ieceiving paiental help.
Not suipiisingly, given what we have seen with the American Time Use Survey,
TV-watching occupied the most weekly leisuie time of 6- to 17-yeai-olds in the Uni-
veisity of Michigan study. (Among non-leisuie activities, only sleep and school atten-
dance claimed moie houis pei week than TV-watching.
18
) Table 3E bieaks down
weekly aveiage houis andioi minutes that 6- to 17-yeai-olds spent on a iange of
leisuie activities foi the 20022003 study peiiod.
Table 3E. Weekly Average Hours and/or Minutes Spent on Various Activities
by American Children
Ages 617, 20022003
Leisure activity 20022003
Visiting, socializing 4:47
Sports 2.59
Outdoor activities 0:50
Hobbies 0:12
Art activities 0:48
Television 14:36
Other passive leisure 2:46
Playing 8:05
Reading 1:17
Being read to 0:05
Computer activities 2:45
Source: University of Michigan, Ann Arbor, Institute for Social Research
Table 3D. Percentage of Daily Leisure Time Spent Watching TV or Reading
Percentage spent watching TV Percentage spent reading
Weekdays Weekends Weekdays Weekends
and holidays and holidays
Total, 15 years and over 51.8% 48.7% 7.3% 6.9%
15 to 24 years 41.9% 38.9% 2.6% 2.6%
25 to 34 years 52.7% 48.6% 4.1% 3.2%
35 to 44 years 52.8% 49.6% 5.6% 4.9%
45 to 54 years 54.1% 50.7% 7.4% 6.7%
55 to 64 years 54.2% 52.0% 10.5% 9.4%
65 years and over 55.3% 52.5% 11.6% 14.0%
Reading and TV-Watching
Why focus on TV-watching time in a report on voluntary read-
ing habits? Admittedly, there is no single barrier, which, if removed, would
raise reading rates for young Americans. Moreover, there is no assurance that a
decline in one type of leisure habit would spur greater participation in another.
We can, on the other hand, use statistics from the 2002 Survey of Public Partici-
pation in the Arts (SPPA) to gauge the strength of the correlation between leisure
reading and TV-watching.
Although the SPPA survey revealed that readers of literature watch the same
average amount of TV per day as non-readersroughly 2 hoursthe likelihood
of reading diminishes with additional hours watched. According to a regres-
sion analysis, the odds of literary reading by Americans who watch 3 or more
hours of TV are significantly less than (or 13% below) the odds of literary reading
by Americans who watch 2 hours or less per day. For all types of book reading
not only fiction, plays, or poetrynon-readers watch an average of 3 hours per
day, while Americans who read at least one book per year watch, on average, 2
hours daily.
As Reading at Risk concedes, the SPPA results cannot show whether people
would read more if they watched less TV, or whether they would use this extra
time in other ways. Elsewhere, the report adds: television does not seem to be
the culprit in the 10- and 20-year declines of American literary reading rates. It is
not the culprit, perhaps, but certainly a culprit in subtracting time and work from
the development of young readersespecially when we regard the SPPA statis-
tics alongside American time-use data and recent medical articles. (See, for
example, Jeffrey Johnson, et al., Extensive Television Viewing and the Develop-
ment of Attention and Learning Difficulties During Adolescence, in the May
2007 Archives of Pediatrics & Adolescent Medicine.)
In their comprehensive study of U.S. time use in the latter part of the 20th
century, Time for Life: The Surprising Ways Americans Use Their Time (1997),
John P. Robinson and Geoffrey Godbey conclude: The real adversary of the
artsis television. Amusing Ourselves to Death: Public Discourse in the Age of
Show Business (1985), Neil Postmans classic polemic against popular media
culture, is more dire and less empirically driven, but its metaphor still resonates:
We are now a culture whose information, ideas and epistemology are given
form by television, not by the printed word. To be sure, there are still readers
and there are many books published, but the uses of print and reading are
not the same as they once were; not even in schools, the last institutions
where print was thought to be invincible. They delude themselves who
believe that television and print coexist, for coexistence implies parity. There
is no parity here. Print is now merely a residual epistemology, and it will
remain so, aided to some extent by the computer, and newspapers and mag-
azines that are made to look like television screens. Like the fish who sur-
vive a toxic river and the boatmen who sail on it, there still dwell among us
those whose sense of things is largely influenced by older and clearer
waters. (p. 28)
19
Data piovided by HSSSE,
Indiana Univeisity Bloomington.
Te 2004 and 2005 suiveys also
asked students how much time
they spent on peisonal ieading
onlineiweb. In 2004, 71 of stu-
dents iead online foi 03 houis
pei week, 17 iepoited ieading
online foi 46 houis pei week,
and the iemaindei iead online foi
7 oi moie houis pei week. In 2005,
the shaie of online ieadeis in the
03-houis categoiy iose to 84.
Still moie iecent data suggest
that online ieading does not con-
tiibute gieatly to the oveiall iead-
ing iates of high school students.
In the 2007 HSSSE, online aiti-
cles was included as an option
in the suivey question about
ieading foi youiself. Of college
fieshmen and seniois alike, moie
than half said they spent only 1
houi pei week, less, oi no time at
all on this activity. (Complete 2007
HSSSE iesults weie unavailable at
the time of publication.)
20
Ibid.
21
Ibid.
Reading accounts foi 1 houi and 17 minutes of the weekly aveiage leisuie time of 6-
to 17-yeai-olds. Tat amount iepiesents less than 9of theii TV-watching time, which
aveiages 14 houis and 36 minutes pei week. But the most ievealing obseivations about
childien and adolescent ieading patteins owfiomsubgioup compaiisons.
Te Univeisity of Michigan study lists aveiage weekly time expendituies foi foui
age ianges within the 6- to 17-yeai-old gioup: 6 to 8, 9 to 11, 12 to 14, and 15 to 17.
Table 3F signals a disciepancy in the leisuie ieading patteins of oldei teenageis, com-
paied with those of youngei age gioups. Foi young childien (6- to 8-yeai-olds), many
of whom aie ielatively new ieadeis, the aveiage weekly time spent ieading is 1 houi
and 22 minutes. Tat numbei incieases foi the 9-to-11 age gioup. Among 12- to 14-
yeai-olds, the time allotment is slightly less. But the aveiage ieading time of 15- to
17-yeai-olds is maikedly dieient.
Tese oldei teens iead foi just undei 50 minutes a weeknot even the full houi
gianted by all the othei age gioups. Taken togethei with what we leained in Chaptei
One about leisuie ieading tiends foi 17-yeai-olds, the Univeisity of Michigan time-
use data suggest a decline in voluntary reading fromchildhood to adolescent age
groups.
Anothei study, Indiana Univeisitys High School Survey of Student Engagement
(HSSSE) 2006, polled 81,499 students of all high school giades fiom 110 schools
acioss the U.S. Responding to questions about vaiious activities, 56 of high school
students attiibuted only 01 houi pei week to ieading foi self. An additional 30
iead 25 houis a week.
As with the fedeial American Time Use Survey, tiend data aie limited foi the pui-
pose of accuiate yeai-ovei-yeai compaiisons. Still, when a similai question was asked
in the 2005 veision of the high school suivey, 54 iepoited peisonal ieading of
books, magazines, newspapeis, etc. foi 01 houis a week. Anothei 34of students
iead those mateiials 2-5 houis weekly foi pleasuie.
19
In 2004, the high school suivey asked iespondents whethei they gave 0 houis, 13
houis, 46 houis, 710 houis, oi 1114 houis to weekly peisonal ieading of books,
magazines, newspapeis, etc. Te majoiity by fai (70) iepoited ieading those mate-
iials foi 03 houis a week. Yet in 2005, a gieatei peicentage of high school students
(77) weie in the 03 houis categoiy.
20
In Chapteis One and Two, we asked whethei homewoik is ciowding out students
leisuie time devoted to voluntaiy ieading. Fuithei data fiomthe High School Survey
of Student Engagement appeai to dispel this notion. In 2004, 2005, and 2006, the
majority of high school students read only minimally for class. In 2004, as we
have seen, 70 spent 03 houis weekly on leisuie ieading. Tat yeai, 76 of high
school students iepoited ieading assigned texts oi couise mateiials foi only 03
houis a week.
21
Te following yeai, that guie was 81.
In 2006, the question on compulsoiy ieading was phiased dieiently. Students
weie asked about the amount of time ieadingistudying foi class, and they selected
Table 3F. Weekly Average Hours and/or Minutes Spent Reading by American
Children, By Age: 20022003
68 911 1214 1517
1:22 1:33 1:25 0:49
Source: University of Michigan, Ann Arbor, Institute for Social Research
22
HSSSE, Indiana Univeisity
Bloomington, Voices on Student
Engagement: A Report on the 2006
High School Survey of Student
Engagement, 2007.
23
Pew Inteinet & Ameiican Life
Pioject, Teens and Technology,
2005, 22.
24
Steve Lohi, Slow Down, Biave
Multitaskei, and Dont Read Tis
in Tiac, e New York Times,
Maich 25, 2007.
25
Shamsi Iqbal, Univeisity of Illi-
nois, and Eiic Hoivitz, Miciosoft
Reseaich, Disiuption and Recov-
eiy of Computing Tasks: Field
Study, Analysis, and Diiections
(San Jose, Calif.: CHI 2007).
fiom ve possible choices: 0 houis, 1 oi fewei, 25, 610, and 10 oi moie. Fifty-ve
peicent selected 0-1 houis.
22
Yet as shown above, a compaiable peicentage (56) also
iepoited ieading foi self foi 01 houis.
Generation M: Reading While Doing Other Activities
Ameiicans laigely accept the iole of multitasking in theii daily ioutines at home and
at woik. Tey may even view the ability to accomplish seveial disciete tasks, simul-
taneously, as a souice of piide in theii eciency. Economic piessuies and time con-
stiaintsthe one factoi scaicely distinguishable fiom the otheihave combined
with incieasingly usei-fiiendly technologies to cieate a scenaiio wheie it no longei
seems unnatuial to talk on the phone while diiving, to giab lunch while Web-suing,
oi to compose an oce memo while a 24-houi TV news channel blaies in the back-
giound.
Eaily data suggest that the use of some types of digital media foi multitasking is
moie pievalent among youth.
23
Much moie ieseaich is needed to weigh the ielative
costs and benets of IT multitasking. Neveitheless, a Maich 2007 New York Times
aiticle posed the question, citing seveial published and unpublished studies by neu-
ioscientists, psychologists and management piofessois.
24
Foi example, the aiticle
iefeiences a eld study of Miciosoft employees woik habits.
Te study showed that aftei inteiiupting theii computei tasks to iespond to e-mail
oi instant messages, employees took, on aveiage, an additional 1015 minutes to
iesume woik. In 27 of the cases, the inteiiuptions iesulted in employees taking
moie than two houis until task iesumption.
25
Te Times quotes the studys co-authoi,
a Miciosoft ieseaichei, as saying: If its this bad at Miciosoftit has to be bad at
othei companies, too.
Taking time o in the middle of the woikday to iespond to a peisonal e-mail is one
type of ineciency, but a mattei best left foi the employei to iesolveif the costs aie
signicant enough to meiit attention in the ist place. By contiast, theie is no easy
method to account foi the costs incuiied by ITmultitasking while attempting to iead
a book. Still, while we may lack substantive data on any tiadeos associated with this
behavioi, we do know that the tendency is iampant among Ameiican youth.
In late 2006, the Heniy J. Kaisei Family Foundation issued the iepoit Media
Multitasking Among American Youth: Prevalence, Predictors and Pairings. Te iepoit
built on an eailiei study by the foundation. Titled Generation M: Media in the Lives
of 8-18 Year-Olds, the eailiei study iepoited, among othei ndings, that young peo-
ple use multiple media 26of the time that they use any media at allwhethei listen-
ing to music, using the computei, watching TV, oi ieading. Tis peicentage maiks a
10-point climb fiom1999, when 16of childien and adolescents iepoited multime-
dia use.
Te Generation M study iesults came fiom a nationally iepiesentative sample
exceeding 2,000 students in giades 312. In addition, the study involved 694 weekly
diaiies on media use. Tese data allowed the Kaisei Family Foundation to analyze
youth multitasking foi the 2006 study, Multitasking Among American Youth.
Tables 3G and 3H display two of the analysis coie ndings: 58% of 7th- to 12th-
graders multitask while reading, some oi most of the time. To considei multitask-
ing in othei teims: 35 of total weekly ieading time foi childien and teens is shaied
with othei media. Scieen media alonevideoicomputei games, e-mail oi instant
messaging, Web suing and, of couise, TV-watchingaccount foi 20 of ieading
time, accoiding to the analysis.
26
David Touve and Steven Teppei,
Cuib Centei foi Ait, Enteipiise
and Public Policy, Vandeibilt
Univeisity, Leisure in America:
Searching for the Forest Amongst
the Trees, 2007, 1718.
Again, moie data aie needed to show conclusively that multitasking, especially
with infoimation technology, helps oi hindeis pioductivity in any single task. But
given what we know about ieading as a sustained act of paiticipation with a text, an
act iequiiing gieat iesouices of memoiy, imagination, and intent questioning, it
seems unlikely that multiple diveisions during the reading process itself can do any-
thing but dilute the ieadeis expeiience and enjoyment of a liteiaiy woik.
Add to this possibility that the dilution is occuiiing at the foimative stages of a
ieadeis developmentwhen any ieading at all competes heavily with othei demands
on a youths leisuie timeand the piospect giows moie tioubling. In a iecent white
papei, Vandeibilt Univeisity ieseaicheis David Touve and Steven Teppei posit that
the multitasking way is heie to stay:
Leisuie is fully embedded within the context of eveiyday ioutines, woik and
social obligations. We may expect those who oei leisuie choices to t into this
polychionic, punctuated, even inteiiupted lifestyle.
26
In the next chaptei, we will considei how book buying has been aected by this
evei-expanding, sometimes deafening aiiay of leisuie choices.
Table 3H. Percentage of Time Spent Reading While Using Other Media
Percentage of reading time
Reading while:
Watching TV 11%
Listening to music 10%
Doing homework on the computer 3%
Playing videogames 3%
Playing computer games 2%
Using the computer (other) 2%
Instant messaging 2%
E-mailing 1%
Surfing websites 1%
Using any of the above media 35%
Source: The Henry J. Kaiser Family Foundation, Media Multitasking Among Youth: Prevalence, Predictors and
Pairings, (# 7592), 2006
Table 3G. Percentage Using Other Media While Reading
% who use other media while reading
Most of the time 28%
Some of the time 30%
Most/some 58%
Little of the time 26%
Never 16%
Little/never 42%
Source: The Henry J. Kaiser Family Foundation, Media Multitasking Among Youth: Prevalence, Predictors and
Pairings, (# 7592), 2006
Table 3I. Book-Reading Rates as Measured by Multiple Surveys
Survey Surveying Number of Response Survey
Year Organization Respondents Rate Question Result
2002 U.S. Census Bureau 17,135 adults 70% With the exception of 57%
and the National books required for work or
Endowment for the Arts school, did you read any books
during the last 12 months?
2005 The Gallup Organization 1,006 adults 20% Do you happen to be reading *47%
any books or novels
at present?
2005 Mediamark Research, Inc. 9,823 adults 38% Checklist of leisure activities, 35%
including reading books,
under Participated in the last
12 months.
2006 Pew Research Center 3,204 adults plus 29% for landline, Not including school or 38%
for the People & the 250 cell-phone-only 24% for cell- work-related books, did you
Press (with Princeton respondents phone only spend any time reading a book
Survey Research yesterday?
Associates International)
2007 Associated Press/Ipsos 1,003 adults N/A Have you read any books in *73%
the past year or havent you
had the chance to read a book
in the past year?
* Does not exclude reading required for work or school.
Different Questions, Different Answers
Since 2004, when the Arts Endowment published its Reading at Risk report,
several organizations have conducted surveys of book-reading by the general
adult population. The surveys vary by sample size, methodology, questions
asked, and data reported. For that reason, specific characteristics of each survey
are provided with the results below.
27
Also, while the peicentage of
45- to 54-yeai-olds in the geneial
population giew foui points fiom
1991 to 2004 (fiom 10.2 to
14.2), its shaie of the U.S. book-
buying maiket iemained at.
(Population data foi the age
gioups deiive fiom the U.S.
Census Buieau.)
28
NEA, Reading at Risk: A Survey
of Literary Reading in America,
2004, xi.
29
Bowkei piess ielease, May 9,
2006.
30
Albeit N. Gieco, e Book Pub-
lishing Industry, 2nd ed. (Mahwah,
N.J.: Lawience Eilbaum Associ-
ates, 2005), 341.
31
In contiast to the boom in title
output, book sales have been chai-
acteiized as sluggish. New book
pioduction jumped 64 between
99 and 04, but sales incieased by
baiely 10, the industiy tiade
magazine Publishers Weekly
iepoited in 2005 (Cianking It
Out, May 30).
CHAPTER FOUR
6. American families are spending less on books than at almost any other time
in the past two decades.
BACKGROUND
W
ith the 21st centuiy baiely begun, the woid matuie can desciibe both
the U.S. book maiket and its customei base. In 1991, Ameiicans in the
25- to 44-yeai-old age gioup accounted foi about 42of the book maiket
(excluding childiens books). Tiiteen yeais latei, they iepiesented only 28. Ovei
the same peiiod, the peicentage of book buyeis in the 55-and-oldei categoiy leapt
13 points to 44. See Table 4A.
Not suipiisingly, the aging of Ameiicas baby boomeisa demogiaphic shift that
has alteied business piactices acioss the commeicial sectoi, in elds such as health-
caie, tiavel, insuiance, and tianspoitationhas also aected book buying and pub-
lishing tiends. Howevei, the peicentage shifts in Table 4A do not unifoimly ieect
changes in the age of the geneial population.
Foi example, fiom 1991 to 2004, the shaie of 25- to 34-yeai-olds nationwide
slipped by two peicentage pointsfiom 16 to 14and this gioups stake of the
total book-buying maiket also decieased, fiom19to 11. But the 3544 age gioup
exhibited a dieient tiend: while the gioup held laigely steady as a peicentage of the
total population, it too made up a signicantly lowei shaie of book puichaseis in
2004, compaied with the 1991 level.
27
Beyond the matuiing of baby boomeis, then, we must look elsewheie to explain
the decline in 25- to 44-yeai-olds as a peicentage of the book-buying public. Pait of
the ieason exists in theii leisuie ieading patteins ielative to othei age gioups.
Te National Endowment foi the Aits 2004 Reading at Risk iepoit ievealed that
the demogiaphic gioup with the highest peicentage of liteiaiy ieadeis (ieadeis of
novels, shoit stoiies, plays, oi poetiy) was no longei young adults. In 1982, Ameii-
cans ages 1834 weie the most likely to iead liteiatuieat a iate of about 60.
Twenty yeais latei, howevei, 45- to 54-yeai-olds had the highest liteiaiy ieading iate:
52.
28
Te study also found a foui-point decline in the peicentage of Ameiicans who iead
any book at all within the past 12 months: 61 in 1992 veisus 57 in 2002. Behind
this tiend weie substantial declines in the book ieading iates of 18- to 44-yeai-olds,
as noted eailiei in this iepoit (see Table 1A on p. 27).
Table 4A. Percentage of Book Purchasers, by Age
1991 2004
Under 25 years old 4.4% 5.1%
25 to 34 years old 18.7% 11.0%
35 to 44 years old 23.7% 17.4%
45 to 54 years old 22.4% 22.8%
55 to 64 years old 15.6% 21.0%
65 years and older 15.2% 22.8%
Source: Ipsos-NPD, reported in the U.S. Census Bureaus Statistical Abstract of the United States, 2003 (Table
1138) and 2006 (Table 1128)
32
Industiy Study Gioup, 2007), vaii-
ous pages. Total book sales include
data foi the following categoiies:
adult and juvenile tiade, mass-
maiket papeiback iack size, book
clubs, mail-oidei publications,
ieligious, piofessional, univeisity
piess, elementaiy-high school,
college, and subsciiption iefei-
ence, excluding standaidized tests.
33
Te annual iepoiting schedules
vaiy foi Amazon, Baines & Noble,
and Boideis.
34
Baines & Noble, Inc., 2006
Annual Report, 2, and Boideis,
Annual Report 2006, Shaieholdeis
Lettei.
35
Te Ameiican Bookselleis Asso-
ciation, which iepiesents inde-
pendently owned bookstoies,
diaws attention to a 6.6 dip in
bookstoie sales in June 2007, com-
paied with the piioi-yeai peiiod.
Te gioups newslettei, Bookselling
is Week (Aug. 13, 2007), cites
U.S. Census Buieau estimates of
$1.13 mil. in ietail bookstoie sales
foi June 2007, compaied with June
2006 sales of $1.21 mil. Tis was
the twelfth month in a iow that
book sales failed to keep pace with
the pievious [yeais] iesults, the
newslettei states.
Retail bookstoie sales tiend data
foi 19922005 aie available
thiough the Census Buieaus
Annual Retail Tiade Suivey. Ovei
the 13-yeai peiiod, ietail book
sales giew 49 to $8.45 mil., aftei
adjusting foi ination. A steady
giowth pattein was undeicut
aftei 2000, with declines in 2001
(-1.2), 2002 (-0.1), and 2005
(-0.9), and incieases in 2003
(3.3) and 2004 (1.8).
36
Unit sales aie computed by
dividing net publishei ievenues by
the wholesale piice of books. Net
publishei ievenue numbeis come
fiom publisheis quaiteily and
annual iepoits and confeience
calls. Bowkei piovides the whole-
sale piice numbeis. All unit sale
numbeis aie net, thus ieecting
total sales aftei ietuins.
37
vaiious pages.
Industry Book Sales Figures
How aie these declines being iegisteied by the book industiy? On the suiface, theie
is no connection. In 2005, U.S. publisheis biought out 172,000 new titles and edi-
tions.
29
Tat total iepiesents an 18,000-unit diop fiom 2004the ist decline in
annual title output since 1999, accoiding to the industiy data publishei Bowkei.
Nonetheless, the 2005 amount is almost thiee times the title output of a decade eai-
liei: 62,000.
30
Publishei ievenues also have giown, albeit at a slowei iate.
31
In 2006, publisheis
did $28.6 billion in total book sales (aftei ietuins). Tis amount is $800 million up
fiomthe pievious yeai, and a $3.9 billion inciease ovei the 2000 level. Sales fiomcon-
sumei books, which aie a subset of all books and include only adult and juvenile tiade
books and mass-maiket papeiback iack-size books, contiibuted to this iise. Net con-
sumei book sales weie $9.2 billion in 2006, up $1 billion fiom 2000.
32
Likewise, sales guies fiom the nations laigest bookselleis aie iobustthough in
many cases it is dicult to ciedit books alone. In 2006, Amazons net sales foi
Media items (including DVD ientals, DVDivideo and music pioducts, magazine
subsciiptions, softwaie games and video consoles, as well as books) weie $7.1 billion,
a 20 climb fiom the pievious yeai.
Foi less diveisied bookselleis, 2006 net sales weie at oi slightly positive: Baines
& Noble and Boideis biought in $5.3 billion and $4.1 billion, iespectively. (Bottom-
line numbeis foi all thiee companies give a moie nuanced pictuie: In 2006, Baines &
Nobles net income iose 3 above the 2005 level, Amazons declined by neaily half,
and Boideis posted a net loss.)
33
Like theii client base, howevei, the maiket foi bookselleis and publisheis is matuie.
Te industiy is peipetually chasing the next hot title oi maiket segment to achieve
the high-giowth potential of othei businesses in the enteitainment sectoi.
Despite Baines &Nobles lofty sales guie, the companys chaiiman acknowledged
in a lettei to shaieholdeis that 2006 was a challenging yeaias a lack of high-piole
titles contiibuted to softness acioss the bookselling industiy. Te CEO of Boideis
Gioup sepaiately ielayed to shaieholdeis his concein that the booksellei has not
moved fast enough to keep up with iapid changes in howconsumeis access infoima-
tion and enteitainment.
34
Both laige booksellei chains, and seveial otheis, supplement book sales with the
maiketing of coee, CDs and DVDivideo pioducts, magazines, stationeiy, and othei
items. Tose additional types of ievenue aie captuied in the net sales data iepoiting
foi both ims. Foi small, independent, and specialty bookstoies, a slump in yeaily
book sales can be debilitating, having a palpable eect on the neighboihood and com-
munity, wheie the stoie may piovide the only locus foi liteiaiy event piogiamming,
book clubs, and authoi ieadings.
35
Consumer Book Unit Sales
Bettei than total sales guies per se, book unit sales aie useful in quantifying book
puichases.
36
Foi example, while publisheis ievenues foi consumei books iose fiom
2000 to 2006, the numbei of consumei book units sold fell.
37
Gains in publishei iev-
enues, theiefoie, would have been diiven by highei book piices, not expanded book-
buying.
Chait 4B shows tiend data foi publishei unit sales of consumei books, fiom 2000
to 2006. Amid yeai-to-yeai uctuations, consumei book unit sales climbed thiough-
38
In 2000, unit sales of juvenile
tiade haidback books iose 8,
while juvenile tiade papeiback
book unit sales giew neaily 12,
accoiding to unpublished data
fiom the Institute foi Publishing
Reseaich, Inc.
39
Scholastic piess ielease, Aug. 2,
2007.
40
vaiious pages.
41
Geneial tiade piint books
exclude childiens books, audio,
and digital puichases.
42
Dan Cullen, National Figuies
Show Independents Hold Maiket
Shaie in a Flat Maiket, Book-
selling is Week, May 1, 2003.
43
vaiious pages.
out the 1980s and eaily 1990s, but have since slowed.
In 2000, moie than 1.6 billion consumei books weie puichased in the U.S.a
iecoid high. One possible ieason foi the spike in unit sales is the Haiiy Pottei seiies:
J.K. Rowlings fouith book was ieleased in haidback in Januaiy 2000, and a papeiback
veision of the second book became available in August of that yeai.
38
Residual sales fiom publication of the ist thiee Pottei books, which came out in
1998 and 1999, also may have contiibuted. Since 2001, no single title has helped to
iestoie unit sales to the 2000 level. (It iemains to be seen whethei Rowlings nal
Haiiy Pottei book, issued in 2007, will do the tiick. Te seiies publishei, Scholastic
Coip. iepoited 8.3 million haidback copies sold in the ist 24 houis of its ielease.
39
)
By 2006, unit sales foi consumei books weie down 100 million fiom 2000, with 1.5
billion books sold.
40
Puichasing tiends foi geneial tiade piint books ieect a similai slowing.
41
Unit
sales giewbetween 1992 and 1996, but tiailed o neai the tuin of the centuiy, baiely
keeping pace with population giowth, accoiding to an industiy consultant quoted in
Bookselling is Week, the newslettei of the Ameiican Bookselleis Association, iep-
iesenting independent bookstoies.
Fiom 19922002, unit sales of geneial adult tiade piint books giew 2.4, but foi
the last 5 yeais of that peiiod, the giowth iate was only 1.3ioughly the same as
foi the U.S. population, the newslettei iepoits.
42
Teie aie no signs that the U.S. book maiket will pick up diamatically anytime
soon. Between 2001 and 2011, the compound annual giowth iate foi consumei book
sales is piojected to be only 0.2.
43
Tese backgiound statistics aie intended to showhowcomplex and multi-layeied
must be any analysis of the book-buying maiket that seeks to quantify the link
between leisuie ieading declines and the health of the commeicial publishing indus-
tiy. Instead of puisuing this aim, which exceeds the scope of oui iepoit, the iemain-
dei of Chaptei Foui will iely on a laige fedeial data souice to examine book-buying
in ielation to othei consumei spending patteins:
1,800
1,600
1,400
1,200
2000 2002 2004 2006
(
i
n
m
i
l
l
i
o
n
s
)
Chart 4B. Unit Sales of Consumer Books
The data represent publishers unit sales net of returns.
Source: Book Industry Study Group, Book Industry TRENDS, 2007
44
Foi the puipose of this analysis,
family andioi household is
used instead of the Buieau of
Laboi Statistics technical teim
consumei unit. In addition to
families and households, a con-
sumei unit may desciibe a pei-
son living alone oi shaiing a
household with otheis oi living as
a ioomei in a piivate home oi
lodging house oi in peimanent
living quaiteis in a hotel oi motel,
but who is nancially independ-
ent. See Fiequently Asked
Questions on the BLS Web site
(http:iiwww.bls.govicexi
csxfaqs.htm).
45
Unpublished data piovided by
the Buieau of Laboi Statistics.
46
Anindya Ghose, New Yoik Uni-
veisity, et al., Inteinet Exchanges
foi Used Books: An Empiiical
Analysis of Pioduct Cannibaliza-
tion and Welfaie Impact, Infor-
mation Systems Research, Maich
2006, 319.
Depaitment of Labois Buieau of Laboi Statistics: Consumer Expenditure
Survey, 19852005.
Annual Consumer Spending on Books
Te Consumei Expendituie Suivey, administeied by the U.S. Census Buieau foi the
Buieau of Laboi Statistics, collects infoimation about Ameiican household and fam-
ily buying habits, including expendituies, income, and othei chaiacteiistics.
44
Conducted annually since 1980, the suivey consists of two paits: a quaiteily intei-
viewand a diaiy (oi iecoidkeeping) suivey. Te inteiviewfeatuies questions on laige
expendituies such as puichases of piopeity and cais, as well as expendituies that
occui on a monthly basis, such as utility payments. Te diaiy pait is intended to cap-
tuie weekly expendituies on smallei, fiequently puichased items such as food and
beveiages, housekeeping supplies, peisonal caie pioducts, and medications.
Results fiom the suivey indicate that the aveiage annual amount that Ameiican
families spend on books is neai its 20-yeai low.
45
Aftei adjusting foi ination, aveiage
consumei spending on books in 1985 was $33.25. By 2003, it had diopped to $27.69,
and though it gained slightly in each of the next two yeais, by 2005 the amount iep-
iesented a 14 decline ovei two decades. See Table 4C and Chait 4D. Notably, the
shaipest deciease occuiied aftei 2000, when unit sales staited to sag.
Some of the decline in spending may be attiibuted to the giowing populaiity of
used books, especially thiough online sales. On the othei hand, a iecent study has
suggested that most puichaseis of used books thiough Amazon aie committed iead-
eis who also buy newbooks, and that used books aie a means by which ieadeis often
get intioduced to new titles and authois.
46
Reading as Entertainment
Anothei way to undeistand tiend data on book puichasing is to ielate those house-
hold expendituies to othei types of spending. Books aie classied as ieading mate-
iials within the suiveys Enteitainment categoiy, which includes a vaiiety of
potential goods and seivices, ianging fiom compact discs and musical instiuments
to pets, toys, and playgiound equipment. It is useful to compaie aveiage consumei
spending on ieading mateiialsnewspapeis, magazines, and newsletteis, in addition
to bookswith spending on othei enteitainment options.
Table 4C. Average Annual Spending on Books, by Consumer Unit
1985 2005 Change
Consumer unit spending on books $37.74 $57.43 +$19.69
(current dollars)
Consumer unit spending on books $33.25 $28.59 -$4.66
(inflation adjusted
1
)
Percentage change in spending, 19852005 -14.02%
(inflation adjusted)
1
Source: Department of Labor, Bureau of Labor Statistics
47
Te Heniy J. Kaisei Family
Foundation, Zero to Six: Electronic
Media in the Lives of Infants,
Toddlers, and Preschoolers, 2003,
4.
48
Consumei spending on
computeis and Inteinet seivices
is categoiized as Miscellaneous
Household Equipment in the
suivey.
Take electionic media, foi instance. In late 2003, the Heniy J. Kaisei Family Foun-
dation iepoited on the peivasiveness of electionic media in eaily childhood devel-
opment. Titled Zero to Six: Electronic Media in the Lives of Infants, Toddlers, and
Preschoolers, the Kaisei study found that childien fiom six months to six yeais old
use scieen media, on aveiage, foi about 2 houis a day, compaied with ieading oi
being iead to, which occupies ioughly 40 minutes of theii aveiage daily time. Scieen
time foi these childien is thiee times the amount of theii ieading time.
47
Te study also showed the omnipiesence of electionic media in childiens living
spaces. Foi example, half the young childien in the study lived in a home with thiee
oi moie television sets, 36 had a TV in theii bedioom, and 49 had a video game
playei. Tiity-foui peicent of young childien lived in homes ieceiving a newspapei
subsciiption, while almost twice as many (63) lived in homes with Inteinet access.
A logical question is how this media satuiation might be ieected in Ameiican
family buying patteins, paiticulaily in ielation to ieading. Although the Consumei
Expendituie Suiveys Enteitainment categoiy does not include computei-based oi
online enteitainment options, the data do covei television, iadios, and sound equip-
menta subcategoiy that includes video games, CD playeis, cable, and satellite
dishes.
48
Books and ieading mateiials aie fundamentally dieient fiomelectionic media in
a vaiiety of ways. Foi the puipose of oui analysis, howevei, the key distinction is that
TVs, video game consoles, and audio equipmentas single unitssimply cost moie
than most books. Unlike electionic media, books aie usually not subject to techno-
logical innovation oi upgiades, and they do not ieect those advances in substantial
piice incieases fiom yeai to yeai. In that sense, we may expect spending on the
TViaudio categoiy to fiequently outpace spending on ieading mateiials.
Fai moie ievealing is to assess both ieading and electionic media as piopoitions of
familyihousehold enteitainment spending, and to compaie those peicentages ovei
time. Chait 4E and Table 4F do just that. As they illustiate, the shaie of Ameiican
families enteitainment budget that goes towaid ieading mateiials is shiinking.
$26
$28
$30
$32
$34
$36
1985 1989 1993 1997 2001 2005
Chart 4D. Average Annual Spending on Books, by Consumer Unit
Adjusted for Inflation
Due to the small size of this spending category, caution should be used in interpreting year-over-year changes.
Between 1995 and 2005, aveiage annual spending on ieading mateiial dipped
almost ve peicentage points as a shaie of total familyihousehold enteitainment
spending. Yet ovei the same time peiiod, aveiage annual spending on TViaudio
equipment giewalmost foui peicentage points to moie than one-thiid of total fam-
ilyihousehold enteitainment spending, as shown in Table 4F.
Given the Kaisei Family Foundation statisticsand the Ameiican time-use suivey
data discussed in the pieceding chaptei of this iepoitit might have been expected
that consumei spending patteins would ieect incieased time spent with electionic
media in compaiison to books and othei ieading mateiials. Combined with the iel-
atively at unit sales of books in iecent yeais, this consumei spending tiend is yet
anothei indicatoi of the loss of ieading time and the iise of scieen time.
40%
35%
30%
25%
20%
15%
10%
5%
0%
1995 1997 1999 2001 2003 2005
TV and audio equipment
Reading materials
Chart 4E. Average Annual Spending on TV/Audio Equipment Versus
Reading Materials as a Share of All Entertainment Spending, by Consumer Unit
Table 4F. Average Annual Spending on TV and Audio Equipment Versus
Reading Materials as a Share of All Entertainment Spending, by Consumer Unit
1995 2005 Change
TV and audio equipment 33.6% 37.2% +3.6 pp
Reading materials 10.1% 5.3% -4.8 pp
From Page to Screen: Newspapers, Blogs, and
Book Reviews
Books are not the only texts in decline. Newspapers, too, have lost ground.
According to annual data from the Audit Bureau of Circulations, average daily
circulation for major newspapers fell 2% to 3% each year from 2005 to 2007. Part
of the reason lies with American youth. David T. Z. Mindich, author of Tuned Out:
Why Americans Under 40 Dont Follow the News (2004), outlines the problem:
In 1972, 46% of college-age Americans read a newspaper every day. Today
its only 21% percent, according to research by the Roper Center for Public
Opinion Researchs General Social Survey. While many point to new
media as the best hope for rekindling interest in news, only 11% of 18- to 24-
year-olds list news as a major reason for logging on. The Internet is a great
source of news for some, but for most it is a great way of avoiding the news,
to be used for e-mail, instant messages, and other personal information.
(Dude, Wheres Your Newspaper? Chronicle of Higher Education, Oct. 8,
2004.)
Those statistics were updated by separate national survey results published in
July 2007. In Young People and News, researchers at Harvards Joan Shorenstein
Center on the Press, Politics and Public Policy found that only 16% of 18- to 30-
year-olds said they read newspapers daily, while 9% of 12- to 17-year-olds said
they did. (By contrast, 35% of Americans over 30 read newspapers daily.) Even
when they did read newspapers, teens and young adults were more likely than
older readers to skim through the news sections, rather than read the stories.
Young peoples lukewarm stance to newspapers extends to news from other
media. Approximately one-fifth of teens and young adults alike said they use the
Internet daily to access news. Yet of those readers, only 32% of teens and 46%
of young adults actively seek news on the Internet, while 65% and 48% just
happen to come across it.
Newspapers, for their part, remain undeterred. Many maintain blogs on their
Web sites, offering readers the chance to comment on reported information. Sim-
ilarly, many news reporters now include their e-mail addresses in their bylines,
encouraging contact with their readership. Perhaps as a result of these changes,
the Newspaper Association of America was able to announce that Web sites of
newspapers had increased their audience by 7.7% in the second quarter of 2007,
compared with the prior-year period. And in January 2007, Nielsen/NetRatings
reported that visits to blogs at the Web sites of the nations top newspapers had
soared 210% year over year in December.
Blogging seems a safe bet for the attraction of teen and young adult readers to
news media. According to Bloggers: A Portrait of the Internets New Storytellers,
a 2006 study by the Pew Internet & American Life Project, 54% of bloggers are
under 30 years old.
But if this interactive capability endows the average reader with a public forum
to discuss news developments, it also can create a sense of vertigo, as the line
between news and hearsay, between reportage and rumor, becomes increasingly
blurred. Sven Birkerts, author of The Gutenberg Elegies: The Fate of Reading in an
Electronic Age (1994), recently described a conflict between blogging and news-
paper book reviews:
Indeed, I would say we are right now at what feels like a point of vital bal-
ance, and those of us involved with literary journalism and book-reviewing
live with the sense of a balance teetering.The implicit immediacy and
ephemerality of post and update, the deeply embedded assumption of
referentiality (linkage being part of the point of blogging), not to mention a
new of-the-moment ethos among so many of the bloggers (especially the
younger ones) favors a less formal, less linear, and essentially unedited
mode of argument. (Lost in the Blogosphere: Why Literary Blogging Wont
Save Our Literary Culture, Boston Globe, July 29, 2007.)
Opinions aside, there is a shortage of scientific research on the effects of
screen readingnot only on long-term patterns of news consumption, but more
importantly, on the development of young minds and young readers. (A good
research question is whether the hyperlinks, pop-up windows, and other extra-
textual features of screen reading can sharpen a childs ability to perform sus-
tained reading, or whether they impose unhelpful distractions.) Some of the
difficulty stems from the constantly evolving nature of information technology,
which often requires equally adaptive research tools to study it effectively. The
news market cannot wait, however, until the definitive study is conducted.
Meanwhile, many newspapers have scaled back their print content, often in
the arts or cultural sections, to make up for lost revenue. Recently, as Birkerts
notes, The Boston Globe, Los Angeles Times, Chicago Tribune, and The Atlanta
Journal-Constitution all have cut or reallocated space for book reviews, a form of
news coverage originally designed to benefit the serious leisure reader.
Gareth Brown/Corbis
CHAPTER FIVE
1. Reading scores for 17-year-olds are down.
2. Among high school seniors, the average score has declined for virtually all
levels of reading.
BACKGROUND
I
n this iepoit so fai, we have seen seveial studies in suppoit of the Aits Endow-
ments 2004 conclusion that ieading iates aie compaiatively lowei in young
adults. But Depaitment of Education and time-use suivey data have shown that
the pioblem extends to even youngei Ameiicans, and that it appeais to stait in the
tiansition fiom childhood to adolescence.
Te subsequent loss of potential ieadeis can be estimated, but not the lost liteiaiy
expeiience. By foifeiting access to a iange of emotions and viewpointsin addition
to veibal, aesthetic, and intellectual pleasuiesthe non-ieadei is impaiiing his oi
hei piospects foi communication thiough the wiitten woid. And while we cannot
assign value to a lost liteiaiy encountei, we can tiack the eiosion of basic skills
teenageis biing to ieading a pieviously unseen text.
Tis discussion will avoid the heavily studied and debated topics of childhood lit-
eiacy and the piopei way to teach ieading. In 1997, Congiess asked the National
Institute of Child Health & Human Development to convene a National Reading
Panel to iecommend ieading instiuctional methods. Composed laigely of psycholo-
gists and educatois, the panel issued a 2000 iepoit that infoimed the Reading Fiist
initiative, legislated by the No Child Left Behind Act of 2001.
Since the laws enactment, debate continues ovei which classioom tools and cui-
iicula woik best to impiove student liteiacy. Tis chaptei will not ievisit claims foi
one technique veisus anothei. Rathei, it assumes a logical, ielatively stiaightfoiwaid
link between ieading often and ieading well, while iecognizing that the link may be
two-diiectional. Foi statistics suppoiting this connection, tuin to Chaptei Seven.
As noted in the ist chaptei of this iepoit, a piepondeiance of infoimation on
teenage ieading tiends is available thiough the Depaitment of Education, which
administeis the National Assessment of Educational Piogiess (NAEP). Te following
data can be found in thiee NAEP iepoits:
NAEP Trends in Academic Progress: ree Decades of Student Performance in
Reading and Mathematics, 2005.
e Nations Report Card: 12th-Grade Reading and Mathematics 2005, 2007.
e Nations Report Card: Reading, 2005.
Src:io To
What the Declines Mean foi Liteiacy
49
See What aie the Dieiences
between Long-Teim Tiend NAEP
and Main NAEP? on Te Nations
Repoit Caid website (http:iinces.
ed.govinationsiepoitcaidiabouti
ltt_main_di.asp).
50
Tey aie Level 200, Demon-
stiate Paitially Developed Skills
and Undeistanding, Level 250,
Inteiielate Ideas and Make Genei-
alizations, and Level 300, Undei-
stand Complicated Infoimation.
9-, 13-, and 17-Year-Olds: Reading Score Trends
Te NAEP long-teim tiends assessment, explained in Chaptei One, ieviewed 2004
ieading compiehension scoies foi 12,00013,000 students in each of thiee age
gioups: 9-, 13-, and 17-yeai-olds. Te nationally iepiesentative study compaied avei-
age ieading scoies foi each gioup with aveiages fiom pievious yeais.
Te test has been conducted 11 times since 1971. It piesents ieadeis with biief pas-
sages of text and a multiple choice foimat asking the ieadei to pinpoint infoimation,
diaw infeiences, and identify the main idea of a passage.
49
A few of the questions
aie open-ended, iequiiing a wiitten iesponse.
Scoiing occuis on a 0- to 500-point scale, with 350 oi highei denoting the ability
to leain fiom specialized ieading mateiials such as scientic, liteiaiy, oi histoiical
essays. At the othei end of the scoiing iange, Level 150 ieadeis can caiiy out simple,
disciete ieading tasks. Tese ieadeis can select woids, phiases, oi sentences to
desciibe a simple pictuie and can inteipiet simple wiitten clues to identify a common
object. Between Levels 150 and 350, theie aie thiee inteimediate ieading levels.
50
Chaptei One ievealed that in contiast to 9- and 13-yeai-olds, 17-yeai-olds iead at
signicantly lowei iates than in 1999 and 1984. Howdo these age gioups faie in iead-
ing compiehension?
Chait 5Adepicts the pattein of aveiage ieading scale scoies foi 17-yeai-olds since
1984. Tat yeai maiked the ist statistically signicant dieience fiomthe 2004 avei-
age scoie foi this age gioup since the test was ist administeied in 1971. As shown by
the chait, aveiage ieading scale scoies weie at foi the thiee test yeais aftei 1984.
Ten the scoies began to slip, iesulting in a ve-point loss fiom 1992 to 2004.
By 2004, the aveiage scale scoie had ietieated to 285, viitually the same scoie as
275
280
285
290
295
300
1984 1988 1990 1992 1994 1996 1999 2004
(Test years occurred at irregular intervals.)
19922004: Down 5 points
285
288 288
288
290* 290* 290*
289*
Chart 5A. Trend in Average Reading Scale Scores for Students Age 17
Test Years 19842004
* Significantly different from 2004.
51
Regaiding the peifoimance
iange of 9-yeai-old ieadeis, the
scoie at the 50th peicentile sui-
passed the median scoie fiom
eveiy othei yeai of the test. Tii-
teen-yeai-olds showed measuiable
dieiences only at the 75th and
90th peicentile scoies, which
exceeded those of 1971, but not
those of 1999. Only 17-yeai-olds
showed no measuiable dieience
in 1971 oi 1999 scoies at any of
the selected peicentiles (10th,
25th, 50th, 75th, and 90th).
in 1971, though not shown in the chait. Foi moie than 30 yeais, in othei woids, 17-
year-olds have not sustained improvements in reading scores. Foi a vivid pictuie
of 17-yeai-old ieading patteins, see Chait 5B, which displays a slowdownwaid tiend
in scoie dieiences since 1984, when the aveiage scale scoie was 289.
A dieient scenaiio applies to 9-yeai-old ieading scoie tiends, also captuied by
Chait 5B. See Chait 5Cfoi moie detail. Fiom19842004, while 17-yeai-old aveiage
ieading scoies weakened, 9-yeai-old ieadei scoies climbed eight points, with the
most diamatic hike occuiiing in 2004. Tat yeai, 9-yeai-olds scoied highei than in
any pievious assessment yeai11 points ahead of the 1971 aveiage scoie.
51
Tis giowth pattein stands in contiast to 17- and 13-yeai-old ieading scoie tiends.
As Chait 5D illustiates, the scoies foi 13-yeai-olds have iemained laigely at fiom
19842004, with no signicant change between the 2004 aveiage scoie and the scoies
fiomthe pieceding seven test yeais. Although not appaient fiomthe chait, the 2004
scoie does iepiesent a signicant impiovement ovei the 1971 aveiagea foui-point
inciease.
At this point, it is woith stiessing that 13- and 17-yeai-olds show piogiessively
lowei voluntaiy ieading iates, fiom 19842004, compaied with 9-yeai-olds. As
Chaptei One noted, in 2004 the teenage gioups showed signicant declines in the
peicentage ieading almost daily, both gioups also had signicant gains in the pei-
centage ieading nevei oi haidly evei foi pleasuie.
Although no causal ielationship can be deteimined, 9-year-olds were the only
age group that maintained an above-50%almost daily reading rate from1984-
2004and they were also the only group to showmajor improvements in read-
ing test scores. Fuithei ieseaich is needed to identify age-specic social andioi
academic factois undeiyling those patteins.
10
-6
-4
-2
0
2
4
6
8
1988 1990 1992 1994 1996 1999 2004
Reported as differences from 1984 reading scores.
Age 17
Age 9
Chart 5B. Trend in Average Reading Scores for Students Ages 17 and 9
Test years occurred at irregular intervals.
Trend analysis based on data from the U.S. Department of Education, National Center for Education Statistics.
219
200
205
210
215
220
225
1984 1988 1990 1992 1994 1996 1999 2004
211*
212*
209*
211* 211*
212*
212*
S
c
a
l
e
S
c
o
r
e
Chart 5C. Trend in Average Reading Scale Scores for Students Age 9
Test Years 19842004
Chart 5D. Trend in Average Reading Scale Scores for Students Age 13
Test Years 19842004
240
245
250
255
260
265
270
275
1984 1988 1990 1992 1994 1996 1999 2004
S
c
a
l
e
S
c
o
r
e
257 257 257
260
258 258
259 259
*Significantly different from 2004.
Source: U.S. Department of Education, National Center for Education Statistics.
52
See What aie the Main
Dieiences between Long-Teim
NAEP and the Main NAEP?
on Te Nations Repoit Caid
website (http:iinces.ed.govi
nationsiepoitcaidiabouti
ltt_main_di.asp).
53
Senioi aveiage ieading scoies
weie also lowei in 2005 than in
1992 acioss all levels of paiental
education.
54
Te peicentage of high school
seniois at the Advanced ieading
level was 5 in 2005, compaied
with 4 in 1992a change of no
statistical signicance.
55
Examples of test passages fiom
the 2005 main NAEP foi high
school seniois include a guide to
the Washington, D.C. Metio tian-
sit system, Maitin Luthei King,
Ji.s Lettei fiom a Biimingham
Jail, Langston Hughes poem,
Hailem, and a shoit stoiy by
Lucienne S. Bloch. Te test used a
multiple-choice foimat and ques-
tions iequiiing shoit oi extended
answeis.
Reading Proficiency of 12th Graders
In addition to the NAEP long-teim tiend assessment, last administeied in 2004, the
Depaitment of Education tests students by giade level (4th-, 8th-, and 12th-giadeis)
on a biennial basis.
Foi 12th-giadeis, this main assessment occuiied most iecently in 2005, with
moie than 21,000 high school seniois evaluated foi ieading compiehension. Te test
featuied longei passages than the tiend assessment, and measuied a iange of ieading
skills, fiom identifying explicitly stated infoimation, to making complex infeiences
about themes, to compaiing multiple texts on a vaiiety of dimensions.
52
Like the
tiend assessment, the 2005 test adopted a 0500 scoie scale, but it also ianked pei-
foimance levels into thiee categoiies: Basic (at least 265), Piocient (at least 302),
and Advanced (at least 346).
Aftei oui ieview of 17-yeai-old ieading test scoies in the long-teim tiend assess-
ment, the 2005 iesults foi high-school seniois may not be suipiisingbut they
should iaise fiesh concein. Te 2005 test found that the aveiage NAEP ieading scoie
foi 12th-giadeis was 286 out of 500. Te scoie was not signicantly dieient fiom
the aveiage scoie foi the pieceding test yeai, 2002. Tis outcome ts with what we
know about tiends in 17-yeai-old ieading scoies, that they often have iemained at
foi seveial test yeais at a time.
But go back to 1992foui tests agoand it becomes cleai that aveiage 12th-giade
ieading scoies have declined by a signicant six points, placing the ieadeis 16 points
shy of the Piocient level. See Chait 5E.
When viewed acioss a iange of peicentile scoies (10th, 25th, 50th, 75th, and 90th),
the declines appeai foi most of the peifoimance distiibution. Students peifoiming at
the 10th, 25th, 50th, and 75th peicentiles all expeiienced signicant declines fiom
1992 to 2005.
53
Only the highest-scoiing students of 2005those at the 90th pei-
centileexpeiienced no declines, compaied with students at theii peifoimance level
in 1992. Conveisely, ieading scoies at the veiy bottomof the 2005 peifoimance iange
declined the shaipest acioss the whole distiibution. See Table 5F.
Tese extieme tiend dieiences foi the highest and lowest peifoiming ieadeis sug-
gest a giowing dispaiity in educational needs of 12th-giade ieadeis. Bad ieadeis aie
ieading woise, and outstanding ieadeis aie baiely meeting the expectations set by
peifoimance levels in pievious yeais.
Anothei way of undeistanding the spiead oi distiibution of 12th-giade aveiage
ieading scoie declines fiom1992 to 2005 is to considei tiend dieiences foi students
peifoiming at oi above the Basic and Piocient ieading levels. Tis analysis shows
that the peicentage of 12th-giade ieadeis at oi above Basic slipped fiom80in 1992
to 73 in 2005, while the 1992 peicentage of ieadeis at oi above Piocient fell ve
peicentage points ovei the same peiiod.
54
Also notable is that the 2005 declines occuiied in all thiee ieading contexts
established by NAEP. Te assessment classies test passages into thiee categoiies:
ieading foi liteiaiy expeiience (such as shoit stoiies, poems, folktales, and biogia-
phies), ieading foi infoimation (e.g., textbooks, newspapei and magazine aiticles,
essays, and speeches) and ieading to peifoima task (including chaits, public tians-
poitation schedules, instiuctions foi games, iecipes, maps, oi insuiance foims).
55
In each of the thiee ieading contexts, 2005 aveiage scoies declined signicantly
fiom1992. But ieading foi liteiaiy expeiience sueied woist: an 11-point deciease,
fiom290 to 279. Independent of the statistical dieience, the decline in this ability is
alaiming because imaginative liteiatuie fosteis peisonal inquiiy and identication
with othei lives, peispectives, and possibilities. Denied basic skills to paiticipate in
this ait foim, young Ameiicans will lose access to a vital pait of theii cultuial heiitage.
4th- and 8th-Grade Reading Proficiency
Te NAEP main assessment in 2005 includes a fai laigei sample of 4th- and 8th-
giadeis than 12th-giadeismoie than 165,000 and 159,000 iespectivelybecause
that yeai the test involved paiticipation of states and selected uiban distiicts foi the
development of sepaiate iepoits.
56
Of all the thiee giade levels tested, 4th-giadeis weie the only students to see a sig-
nicant climb in the peicentage scoiing at oi above Piocient in 2005, compaied
with 1992: 31veisus 29. But the laigei point is just howpooily students in all thiee
giades aie ieading.
56
Te thiee ieading achievement
levels foi 4th-giadeis coiiespond
with the following scoies (in
paientheses): Basic (at least 208),
Piocient (at least 238), and
Advanced (at least 268). Foi 8th-
giadeis, ieading achievement
levels aie: Basic (at least 243),
Piocient (at least 281), and
Advanced (at least 323). Published
iesults of the 2005 NAEP ieading
tests foi 4th-, 8th-, and 12th-
giadeis do not iepoit cumulative
below Basic peicentages at the
national level foi public and pii-
vate schools. Consequently, the
below Basic peicentages in Table
5G weie obtained by subtiacting
the Basic oi above peicentages
fiom 100.
Table 5F. Change in 12th-Grade Reading Scores, by Percentile: 1992 and 2005
Percentile 1992 2005 Change 9205
90th 333 333 0
75th 315 313 -2
50th 294 288 -6
25th 271 262 -9
10th 249 235 -14
All score changes from 1992 are statistically significant.
Chart 5E. Trend in Average Reading Scale Scores for 12th-Graders
Test Years 19922005
275
280
285
290
295
300
1992 1994 1998 2002 2005
S
c
a
l
e
S
c
o
r
e
292*
287
290*
287
286
*Significantly different from 2005.
Table 5Giepoits ieading achievement levels by giade. In 2005, appioximately 36
of 4th-giadeis iead at a level below Basicthe peicentage had not changed signi-
cantly since 1992while 27of 8th-giadeis iead belowBasic. Te peicentage of 8th-
giade ieadeis below Basic lost foui points fiom 1992 to 2005, becoming identical to
the peicentage of 12th-giadeis in that categoiy: 27. Among 8th-giadeis and 12th-
giadeis alike, one in foui students is a below-Basic ieadei.
Foi ieasons that aie uncleai, 4th-giadeis consistently have had the gieatest pei-
centage of below-Basic ieadeis. But wheie 12th-giadeis foimeily had an 11-point
lead ovei 4th- and 8th-giadeis in teims of the peicentage of piocient ieadeis at each
giade level, by 2005 the gap had nownaiiowed to foui points. Reading scoies of 4th-
and 8th-giadeis aie impioving.
57
See Table 5H. At the lowei giades, aveiage ieading
scoies weie two points highei than in 1992, small but signicant incieases. As shown
eailiei, the aveiage ieading scoie of 12th-giadeis has declined by six points.
Viewing the giade-level ieading test iesults alongside the tiend analysis of ieading
scoies by age, we nd the ieading ability of 17-yeai-olds and 12th-giadeis has eithei
declined oi failed to impiove. By contiast, 9-yeai-olds and 4th-giadeis, on one hand,
and 13-yeai-olds and 8th-giadeis, on the othei, have shown signicant gains.
Te dispaiity in ieading skills impiovementbetween the lowest ageigiade and
the highest ageigiademiiiois the gulf in voluntaiy ieading iates of those gioups.
Te deteiioiation in ieading iates and piociency of 17-yeai-olds makes possible a
scenaiio wheie, accoiding to the 2003 National Assessment of Adult Liteiacy, only
5% of high school graduates are Procient readers, as we shall see in the next
chaptei.
57
Results fiom the NAEP 2007
ieading assessments foi 4th- and
8th-giadeis aiiived as this iepoit
was in the piocess of publication.
Te iepoits ieveal that 4th-giade
ieading scoies aie continuing to
impiove: in 2007, the aveiage
scoie foi this giade level (221) was
highei than in any of the pieced-
ing test yeais, and the peicentage
of ieadeis at the Basic and Pio-
cient levels giew by 5 and 4 points,
iespectively, compaied with 1992
levels. (Like the aveiage test scoie,
the peicentage of 4th-giadeis at
the Basic and Piocient levels also
iose signicantly fiom 2005.)
Eighth-giadeis, too, showed an
aveiage ieading scoie impiove-
ment (fiom 260 in 1992 to 263 in
2007) and a slight but signicant
gain in the peicentage ieading at
oi above the Basic level (74),
compaied with 1992 and 2005 lev-
els. Unlike 4th-giadeis, howevei,
8th-giadeis did not see a signi-
cant change in the peicentage of
Piocient ieadeis, compaied with
those piioi test yeais.
Table 5H. Average Reading Scores of 4th-, 8th-, and 12th-Graders
1992 2005 Change
Grade level:
4th 217 219 +2
8th 260 262 +2
12th 292 286 -6
Table 5G. Percentage of 4th-, 8th-, and 12th-Graders, by Reading Achievement Level
Below Basic Basic or above Proficient or above
1992 2005 Change 1992 2005 Change 1992 2005 Change
Grade level:
4th 38% 36% *-2 pp 62% 64% *+2 pp 29% 31% +2 pp
8th 31% 27% -4 pp 69% 73% +4 pp 29% 31% *+2 pp
12th 20% 27% +7 pp 80% 73% -7 pp 40% 35% -5 pp
* Not statistically significant
All percentages are approximate.
Reading and the Gender Gap
Among 12th-graders, boys and girls alike performed worse in
the 2005 main NAEP reading assessment than in 1992. For girls, however, the
average reading score was lower than in 2002, while for boys the score has
remained constant for the past two test years. Still, because male student scores
were eight points lower in 2005 than in 1992, compared with the five-point differ-
ential for females over the same period, the gap between male and female
reading scores has widened. Further, female students outperformed males in
all three reading contextsliterary reading, reading for information, and reading
to perform a task.
This disparity in reading scores between the genders is not restricted to chil-
dren or teenagers: as will be seen in the next chapter, women outscore men on
adult prose literacy tests. What accounts for females leading males in reading
ability? In many of the studies of voluntary reading that have been cited so far,
females have proved more frequent readers than males, and it is hard not to
credit greater exposure to books with playing a role.
The American Time Use Survey, for example, has consistently shown women to
spend more time on leisure reading than men. The Henry J. Kaiser Family Foun-
dations 2005 Generation M: Media in the Lives of 818 Year-Olds study identified
the same split along gender lines in children and teens. And The National Endow-
ment for the Arts Reading at Risk (2004) report found that more than half of
Americas women (55%) read a novel, short story, poem, or play in the preceding
12 months, compared with 38% of men. Yet even for women, this reading rate
was down eight percentage points from 20 years earlier. The gender factor may
play a role in determining reading rates and reading skills, but females are by no
means out of trouble when it comes to declines in both.
Table 5I. Average 12th-Grade Reading Scores by Gender
1992 2005
Female 297 292
Male 287 279
Male-female gap -10 -13
CHAPTER SIX
3. Reading prociency rates are stagnant or declining in adults of both genders
and all education levels.
BACKGROUND
T
he pieceding chaptei demonstiated that 17-yeai-old and high school senioi
ieading scoies have weakened ovei time. Tis nding has obvious implica-
tions foi college piepaiedness, especially since two-thiids of 2008 high school
giaduates, oi 2.2 million Ameiicans, aie likely to enioll that fall as ist-yeai college
students.
58
In 2006, the nonpiot oiganization ACT, which administeis yeaily college
entiance exams, published an analysis of the moie than 12.5 million high school stu-
dents who had taken ACT tests fiom1994 to 2005. ACT concluded: Student ieadi-
ness foi college-level ieading is at its lowest point in moie than a decade, with little
moie than half of all students meeting ACTs benchmaik scoie.
59
Te oiganization also found, astonishingly, that moie students aie on tiack to
being ieady foi college-level ieading in 8th and 10th giade than aie actually ieady by
the time they ieach 12th giade. Tose data deiived fioma compaiison of ACTiead-
ing test scoies foi 352,000 high school students fiom 1998 to 2005.
Chaptei Eight will ieview data fiom employei suiveys assessing the knowledge
gaps of high school giaduates who entei the U.S. woikfoice. But college giaduates
suei similai deciencies, which may have oiiginated with the mediocie ieading
iates and skills many caiiied thiough high school.
Te Depaitment of Education piovides a snapshot of the ieading ability of Amei-
ican adultsthose who have giaduated fiom college, but also those who have not.
Tis time, oui souice is a single study:
National Centei foi Education Statistics: Literacy in Everyday Life: Results from
the 2003 National Assessment of Adult Literacy, 2007.
Adult Prose Literacy Rates: Overview
Te Depaitment of Education gave the 2003 National Assessment of Adult Liteiacy
(NAAL) to 19,200 Ameiicans ages 16 and above.
60
Te NAAL test booklets consisted
of a seiies of open-ended questions, each pieceding a text passage.
Test-takeis weie scoied on a 0500 scale, with foui distinct piose liteiacy levels:
BelowBasic (0209), Basic (210264), Inteimediate (265339), and Piocient
(340500). An Inteimediate ieading scoie coiiesponds with undeistanding modei-
ately dense, less commonplace piose texts as well as summaiizing, making simple
infeiences, deteimining cause and eect, and iecognizing the authois puipose.
61
By contiast, a ieading scoie of Piocient shows the test-takeis ability to iead
lengthy, complex, abstiact piose texts, synthesize infoimation and make complex
infeiences. Accoiding to the NAAL, this liteiacy level includes the ability to compaie
viewpoints fiom two dieient newspapei editoiials, to compaie and contiast the
meaning of metaphois in a poem, and to infei the puipose of an event desciibed in
a magazine aiticle.
58
Depaitment of Education,
National Centei foi Education
Statistics, Projections of Education
Statistics to 2015, and NEA
Reseaich piojections based on
histoiical NCES data.
59
ACT, Reading Between the
Lines: What ACT Reveals About
College Readiness in Reading,
2006. Te benchmaik coiiesponds
with an ACT Reading Test scoie
of 21.
Few tiend data have been pub-
lished iecently foi the othei well-
known entiance exam, the College
Boaids SAT, which integiated a
new wiiting component in 2005.
In August 2007, the College Boaid
iepoited a one-point decline in
the aveiage ciitical ieading scoie
foi 2007 test-takeis, compaied
with the pievious yeai. Otheiwise,
ciitical ieading scoies have
iemained at foi the past seveial
yeais, accoiding to test ocials.
Yet male students, who histoiically
have outscoied females in this test
subject, aie now only two points
ahead of femalescompaied with
a nine-point lead in 2003. As foi
the new wiiting component,
females aie 11 points ahead of
males: 500 veisus 489 (out of 800).
60
Tis population thus included a
youngei cohoit than the NEAs
Reading at Risk iepoit of liteiaiy
ieading habits. Tat study dened
adults as 18 oi oldei. Te NAAL
tested appioximately 18,000 adults
in U.S. households and 1,200 adult
piisoneis.
61
Literacy in Everyday Life, p. 4.
Foi this analysis, we will examine piose liteiacy scoies only. (Te NAAL also tested
foi document and quantitative liteiacy.
62
) In 2003, 43% of test-takers placed at
or belowBasic, with a third of that group in the BelowBasic category. Tis gioup
iepiesents 96 million Ameiicans, with 30 million at BelowBasic, and the iemaindei
at Basic.
63
Te Below-Basic gioup includes the 3 of all test-takeis deemed nonliteiate in
Englisha guie iepiesenting 7 million Ameiicans. Te designation does not
include adults whose language baiiiei pievented themfiomtaking the test, that g-
uie is estimated at 2 of the U.S. adult population.
Compaiing the 2003 iesults with those of the pievious test yeai, 1992, we nd that
the peicentage of piose ieadeis at the BelowBasic, Basic, and Inteimediate levels has
iemained laigely constant. A statistically signicant change does aiise, howevei, in
the peicentage of Ameiican adults who iead at the Piocient level. Tey slipped fiom
15 in 1992 to 13 in 2003. See Table 6A.
Prose Literacy by Education Level
In addition to taking the liteiacy test, NAAL paiticipants answeied a backgiound
questionnaiie that sought infoimation on theii demogiaphics. Responses to the
questionnaiie allowus to undeistand, among othei vaiiables, the educational piole
of test-takeis. While 2003 aveiage piose liteiacy scoies iose with each incieasing level
of education (excluding adults still in high school), the tiend data foi 19922003
show signicant scoie declines in almost all gioups.
64
See Table 6B.
Stiangely, those declines aie steepest among the best-educated gioups. Giaduate
students oi degiee-holdeis in 2003 scoied, on aveiage, 13 points lowei than theii
1992 counteipaits. Othei college giaduates in 2003 weie down 11 points, on aveiage,
compaied with 1992 college giaduates who did not puisue a highei degiee. By con-
62
Document examples include
job applications, payioll foims,
tianspoitation schedules, maps,
tables, and diug and food labels,
while quantitative liteiacy iefeis to
balancing a checkbook, guiing
out a tip, completing an oidei
foim, and deteimining the amount
of inteiest on a loan fiom an
adveitisement. (See Literacy in
Everyday Life, p. 2.)
63
Te 96 million deiives fiom
the total U.S. population iepie-
sented by the 2003 NAAL study:
221,020,000 adults in U.S. house-
holds and 1,340,000 incaiceiated
adults.
64
Exceptions aie adults still in
high school and those with a GED
oi high school equivalency. In
both cases, scoie changes fiom
1992 to 2003 weie not statistically
signicant.
Table 6B. Average Prose Literacy Scores of Adults, by Highest Level of
Educational Attainment: 1992 and 2003
Education level 1992 2003 Change
Vocational/trade/business school 278 268 -10
Some college 292 287 -5
Associates/2-year degree 306 298 -8
Bachelors degree 325 314 -11
Graduate study/degree 340 327 -13
Table 6A. Percentage of Adults at Each Prose Literacy Level: 1992 and 2003
Below Basic Basic Intermediate Proficient
1992 14% 28% 43% 15%
2003 14% *29% *44% 13%
* no statistically significant change from 1992
tiast, the aveiage scoie foi 2003 high school giaduates was 6 points lowei than foi
high school giaduates in 1992.
Despite exposuie to highei leaining in the case of college and giaduate degiee-
holdeis, then, American adults of virtually all education levels are reading less
well than in the previous decade. One may conjectuie that the eiosion of ieading
scoies is paitly due to less cuiiiculai emphasis on ieading at the highei education
levels, but anothei plausible explanation is that adults aie losing the habit of volun-
taiy ieading, which could sustain and ieinfoice theii liteiacy skills.
Te NAAL iepoit notes a paiadox, howevei. Although piose liteiacy scoies
declined foi high school and college giaduates fiom1992 to 2003, the aveiage piose
liteiacy scoie foi the U.S. adult population iemained constant ovei that peiiod. Te
iepoit explains that the declines in aveiage piose liteiacy scoies at eveiy educational
level weie oset by an inciease in the peicentage of adults with postsecondaiy edu-
cation fiom 1992 to 2003, and a coiiesponding deciease in the peicentage that did
not complete high school. Aftei all, compaied with adults who do not complete high
school, adults with postsecondaiy education tend to attain highei piose scoies.
Te aveiage piose ieading scoie was stable fiom 1992 to 2003, and this tiend
applied to all but the veiy best ieadeis. As we have seen, Piocient was the only lit-
eiacy level to expeiience a signicant changea dipin the peicentage of qualifying
ieadeis. Te decline is moie stiiking when viewed as a function of the test-takeis
education status.
As Table 6Cshows, the decline in the percentage of Procient readers is occur-
ring at the highest educational levels. Fiom1992 to 2003, the piopoition of Pio-
cient ieadeis who had attended giaduate school lost 10 peicentage points. Among
college giaduates who had not pioceeded to giaduate school, the peicentage of Pio-
cient ieadeis diopped nine points. Tose ieductions tianslate to a 20 iate of
decline foi adults with giaduate school expeiience and a 22iate of decline foi othei
college giaduates. Vocationalitiadeibusiness school giaduates also expeiienced a sig-
nicant decline in the shaie of Piocient ieadeis: fiom 9 in 1992 to 5 in 2003.
Prose Literacy by Gender
Te dieiences in male and female adult liteiacy levels iesemble the dispaiities in
ieading skills of boys and giils (see p. 62). In 2003, the aveiage piose scoie foi males
Table 6C. Percentage of Adults Proficient in Reading Prose, by Highest Level
of Educational Attainment
Education level 1992 2003 Change Rate of decline
Less than/some high school 1% 1% 0 0
High school graduate 5% *4% *-1 pp *-20%
Vocational/trade/business school 9% 5% -4 pp -44%
Some college 14% *11% *-3 pp *-21%
Associates/2-year degree 23% *19% *-4 pp *-17%
Bachelors degree 40% 31% -9 pp -22%
Graduate study/degree 51% 41% -10 pp - 20%
* No statistically significant change
was a signicant foui points lowei than in 1992: 272 veisus 276. Females expeiienced
no change in aveiage piose scoie ovei this peiiod (277 in both 1992 and 2003),
although because of the decline in male scoies, women scoies in 2003 exceeded mens
by a statistically signicant amountan event that did not occui in 1992.
Piose scoies foi the best female ieadeis also iemained at fiom 1992 to 2003. Yet
the peicentage of males at the Piocient ieading level did diop signicantly, as Table
6D shows.
Taken togethei, the national piose liteiacy data showa lack of impiovement in the
ieading skills of women and a decline in the ieading skills of men. Adults at viitually
eveiy education level exhibited declines in aveiage ieading scoies.
65
Te data also showa slight but signicant decline in the oveiall peicentage of Pio-
cient-scoiing ieadeis, and declines in ieading piociency among college giaduates
and among adults who ieceived vocational tiaining. Although these tiends suggest
the need to expand adult leaining oppoitunities in oui civic institutions and com-
munities, we should not oveilookas cential to that aimthe piomotion of volun-
taiy ieading habits in giown-ups no less than childien.
65
Othei signicant declines
occuiied among 25- to 39-yeai-
olds, whose aveiage ieading scoie
diopped ve points to 283 fiom
1992 to 2003, and among 40- to
49-yeai-olds, who saw an 11-point
score decrease to 282 ovei the
same peiiod. Te peicentage of
25- to 39-yeai-olds in the Below
Basic piose liteiacy categoiy giew
a signicant two points, to 12.
Table 6D. Percentage of Adults Proficient in Reading Prose, by Gender
Men 16% 13% -3 pp -19%
Women 14% 14% 0 pp 0%
Literacy in the Prison Population
Incarcerated Americans, totaling 1.4 million in 2003, composed
an important part of the National Assessment of Adult Literacy. Since the
1992 test, however, several changes occurred in this population that positively
affected literacy scores in 2003.
By the time of the 2003 test, U.S. prisons had gained half a million adult
inmates. On average, this group was older and better educated than the 1992
prisoners; indeed, parents of the 2003 prisoners also tended to be better edu-
cated than parents for the earlier test population. Related largely to those factors,
the average prose literacy for 2003 was higher than in the 1992 group.
Still, prisoners continued to score significantly lower than non-incarcerated
Americans on prose reading tests. The average prose literacy score of prison-
ers in 2003 was 257, while for non-incarcerated adults it was 275. Moreover,
the percentage of prisoners at or below the Basic reading level was 56%, com-
pared with 43% of non-prisoners.
Forty-one percent of prisoners were Intermediate readers and only 3% were
Proficient. Among non-incarcerated Americans, 44% read at the Intermediate
level. This percentage was not significantly different from the proportion of pris-
oners at that reading level. The proficiency rate for non-prisoners, however, was
13%10 points higher than for prisoners.
A disparity in the scores of U.S. household and prison populations appears
even in adults at the highest education level. The average prose score of non-
incarcerated adults who had a postsecondary school education was 302, com-
pared with 282 for prisoners of the same educational background, or one-fifth
of incarcerated adults. Twenty-three percent of non-incarcerated adults who
attended postsecondary school read at the Proficient level, while only 8% of
prisoners of that background did. For adult prisoners and non-prisoners whose
education ended with a high school diploma, average reading scores were
almost identical: 264 and 262, respectively.
Regarding the literacy gap between adult prisoners and non-prisoners, the
Department of Educations Literacy Behind Bars (2007) report concludes: This
would be of somewhat less concern if prison inmates who expected to be
released within 2 years had higher literacy than inmates with more time left to
serve on their sentences, but that was not the case. As the report observes, The
literacy of inmates who are near their expected date of release may be of particu-
lar concern because they will soon need to do such things as rejoin their families
and find a job. Yet the 2003 literacy assessment revealed no difference in aver-
age prose scores of prisoners who still had more than 2 years to serve, and those
to be released within a 2-year period.
CHAPTER SEVEN
4. Reading for pleasure correlates strongly with academic achievement.
BACKGROUND
O
vei the last thiee decades, cognitive psychologists specializing in childhood
development have sought to claiify the ielationship between ieading widely
and ieading well. Much of that ieseaich has been distilled by Anne Cun-
ningham, Univeisity of Califoinia at Beikeley, and Keith Stanovich, Univeisity of
Toiontonotably in a SpiingiSummei 1998 aiticle foi the Ameiican Fedeiation of
Teacheis magazine American Educator. Titled What Reading Does foi the Mind,
the aiticle begins:
Reading has cognitive consequences that extend beyond its immediate task of
lifting meaning fioma paiticulai passage. Fuitheimoie, these consequences aie
iecipiocal and exponential in natuie. Accumulated ovei timespiialing eithei
upwaid oi downwaidthey caiiy piofound implications foi the development
of a wide iange of cognitive capabilities.
Cunningham and Stanovich go on to desciibe the iecipiocal eects of ieading
volume, wheieby, iegaidless of ieading ability, childien who stait ieading foi pleas-
uie at an eaily age aie exposed to exponentially highei numbeis of newwoidsand
a gieatei oppoitunity to develop liteiacy skillsthan childien denied eaily ieading
expeiiences. Because these skills aie associated not just with ieading compiehension,
but also with gieatei cognitive development, the ieseaicheis elsewheie state: We
believe that independent ieading may help explain the widening achievement dis-
paiities between the educational haves and have-nots.
66
Vocabulaiy giowth is peiceived by many to be cential to ieading development, but
what often gets oveilooked is the sheei quantity of woids, concepts, and types of syn-
tactic stiuctuie that ieading piovides. Cunningham and Stanovich cite statistical
studies of piinted texts ianging fiom childiens stoiies to adult books, fiom comic
books to populai magazines, showing that each type of ieading mateiial contains a
fai gieatei aveiage numbei of iaie woids than TVshows oi adult speech.
67
Te impli-
cation is that childien would leain moie vocabulaiy fiom ieading juvenile ction
than fiom watching a piime-time television show oi listening to theii college-edu-
cated paients chat with each othei!
Anothei aiticle summaiizing ieseaich on the link between voluntaiy ieading and
liteiacy iates was commissioned by the Depaitment of Education in the late 1990s.
68
Its authoi postulates:
Voluntaiy ieading involves peisonal choice, ieading widely fiom a vaiiety of
souices, and choosing what one ieads. Aliteiates, people who have the ability to
iead but choose not to, miss just as much as those who cannot iead at all. Indi-
viduals iead to live life to its fullest, to eain a living, to undeistand what is going
on in the woild, and to benet fiomthe accumulated knowledge of civilization.
Even the benets of demociacy, and the capacity to govein ouiselves success-
fully, depend on ieading.
66
Reading Can Make You
Smaitei, by Anne Cunningham
and Keith Stanovich, in Principal,
NovembeiiDecembei 2003.
Repiinted with peimission. Copy-
iight (2003) National Association
of Elementaiy School Piincipals.
All iights ieseived.
67
Te authois iefei to Hayes,
D.P. and Ahiens, M., Vocabulaiy
Simplication foi Childien: A Spe-
cial Case of Motheiese?, Journal
of Child Language 15 (1988): 395
410, and Hayes, D.P., Speaking
and Wiiting: Distinct Patteins of
Woid Choice. Journal of Memory
and Language 27 (1988): 572585.
68
Beinice E. Cullinan, Indepen-
dent Reading and School Achieve-
ment, Ameiican Association of
School Libiaiians School Library
Media Research, Vol. 3, 2000.
Tat last pioposition will iesuiface in Chaptei Nine of this iepoit, but foi now, we
aie conceined with identifying the coiielation between ieading often and ieading
well. Oui ieviewwill covei the most iecent data available in Depaitment of Education
studies pioling youth ieading skills and voluntaiy ieading iates:
NAEP 2004 Trends in Academic Progress: ree Decades of Student Performance
in Reading and Mathematics, 2005.
National Centei foi Education Statistics: NAEP Data Explorer tool, available
via NCES Web site.
Although eailiei chapteis have hinted at a mutual ieinfoicement between ieading
iates and liteiacy skills, the following giaphs make that pattein explicit. Still, because
the data cannot piove a causal ielationship between the two vaiiables, it is woith
iepioducing a disclaimei fiom the pieviously quoted aiticle.
Longitudinal studies that show long-teim eects oi that isolate the exeicise of
liteiacy, howevei, aie missing fiomthe ieseaich on voluntaiy ieading and school
achievement. Such studies might indicate which factois make a dieience in
establishing lifetime ieading habits, and in what inuences ieadeis choice of
ieading mateiial, that peihaps could help us plan eective piogiams.
69
Reading Often and Reading Well
9-, 13-, and 17-Year-Olds: Long-Term Trend Data
Te consistently documented coiielation between voluntaiy ieading iates and highei
test scoies is summed up by a paiagiaph in the NAEP 2004 Trends in Academic
Progress iepoit. Te study assessed the ieading skills of 9-, 13-, and 17-yeai-olds ovei
thiee decades (see Chapteis One and Five).
At all thiee ages, students who indicated that they iead foi fun almost eveiy day
had highei aveiage ieading scoies in 2004 than those who said that they nevei oi
haidly evei iead foi fun. Students at all thiee age levels who said that they iead
foi fun once oi twice a week had highei aveiage scoies than those who nevei oi
haidly evei iead foi fun.
70
Chait 7A displays this pattein foi 17-yeai-oldsthe age gioup least likely to iead
foi pleasuie, as we have seen, and also the only gioup whose ieading scoies have
declined signicantly since 1984.
Although aveiage ieading scoies aie similai foi 17-yeai-olds who iead once oi twice
a week and those who iead once oi twice a month, theie aie laige scoie dieiences
between these ieadeis and the 17-yeai-olds who iead infiequently oi not at all. Te
gulf expands foi infiequent oi non-ieadeis, compaied with those who iead almost
eveiy dayyielding a 33-point gap between daily ieadeis and veiy infiequent iead-
eis, and a 37-point gap between daily ieadeis and nevei-oi-haidly-evei ieadeis.
8th- and 12th-Graders: National 2005 Data
Te NAEP national main assessment, administeied biennially to 4th-, 8th- and
12th-giadeis, ieects the same pattein. Test-takeis aie asked howoften they iead foi
69
Ibid.
70
NAEP Trends in Academic
Progress: ree Decades of Student
Performance in Reading and
Mathematics, 2005, vii.
fun on theii own. Chait 7B shows the 2005 test scoie pattein of 12th-giadeis in foui
dieient categoiies of ieading fiequency. Although not shown heie, the ielationships
between ieading iates and test scoies aie similai in pievious test yeais, and foi all
thiee giade levels.
Anothei question on the 2005 main NAEP assessment asked 8th- and 12th-giadeis
howoften, if evei, they iead ction books oi stoiies outside school. Again, the iesults
aie similaifoi both giade levels, and in pievious yeais. Chait 7C displays the 2005
pattein foi 12th-giadeis. Notably, theie is no signicant dieience in scoies between
the 12th-giadeis who iead only once oi twice a month and those who iead a fewtimes
a yeai. Although this anomaly does not appeai in the pievious two chaits, it ieinfoices
the notion that consistently high levels of leisuie ieading aie ciucial to a positive iela-
tionship between ieading and test scoies.
Almost every day
A few times a year
305
288
287
272
268
Almost every day
302
292
285
274
Chart 7A. Average Reading Scores by Frequency of Reading for Fun
Age 17 in 2004
Chart 7B. Average Reading Scores by Frequency of Reading for Fun
Grade 12 in 2005
Source: U.S. Department of Education, National Center for Educational Statistics
Reading Often and Writing Well
Te National Assessment of Educational Piogiess (also known as Te Nations
Repoit Caid) featuies a wiiting component. Te test assigns students thiee types of
wiiting tasks: naiiative, infoimative, and peisuasive. Assignments fiomthe 2002 test
included, foi example, wiiting a lettei to the editoi of a newspapei, oeiing advice
to youngei students, iepoiting to a school committee, and wiiting a stoiy based on
a poem.
71
Without citing a causal ielationship between the two vaiiablesonce again, data
do not exist to showconclusively that one pattein iesults fiomthe otheiwe note a
piogiessive iise in test scoies by ieading fiequency. Chait 7D depicts a 29-point
spiead between the 12th-giadeis who aie non- oi infiequent ieadeis and those who
iead daily oi almost daily. (Foi 4th- and 8th-giadeis, the coiiesponding dieiences
aie 20 and 25 points, iespectively.)
Wiiting and ieading aie sepaiate skills, to be suie, but the likelihood of wiiting well
is ieduced when one has minimal exposuie to piinted texts. Does the same tendency
apply to ieadeis and non-ieadeis tested foi academic subjects beyond ieading and
wiiting?
Books in the Home
Te NAEP does test foi subjects such as math, science, histoiy, and civics, but stu-
dents aie not asked about theii voluntaiy ieading iates. Rathei, the NAEP inquiies
about the numbei of books in the home. While this vaiiable says nothing of the stu-
dents piopensity foi leisuie ieading, access to printed matter is associated posi-
tively with test scores for a variety of subjects.
Foi example, math, science, civics, and histoiy scoies aie highei foi 12th-giadeis
who iepoit having books in the home. See Tables 7E and 7F, which iepoit, by numbei
of books in the home, the aveiage math and science scoies of 12th-giadeis. Te
iesults aie fiom the most iecent NAEP tests.
71
See What Does the NAEP
Wiiting Assessment Measuie?
on NCES Web site (http:iinces.ed.
govinationsiepoitcaidiwiitingi
whatmeasuie.asp). Data fiom the
2007 NAEP wiiting assessment
weie unavailable at the time of
piepaiing this publication.
At least once a week
Few times a year
302
289
287
273
Chart 7C. Average Reading Scores by Frequency of Reading Fiction Books or
Stories Outside School
Grade 12 in 2005
Te NAEP iesults of a 2006 civics test show a similai pattein, as do iesults fiom a
2006 histoiy test. In contiast to students who iepoited 010 books at home, 12th-
giadeis with 26100 books at home scoied 27 points highei. If they had moie than
100 books at home, the scoie went up an additional 17 points.
Tat same yeai, histoiy test-takeis with 26100 books at home scoied 24 points
moie than 12th-giadeis with 010 books at home. If they had moie than 100 books,
the scoie was anothei 16 points gieatei. But even foi households with 1125 books,
Almost every day
165
154
149
136
Chart 7D. Average Writing Scores by Frequency of Reading for Fun
Grade 12 in 2002
Table 7E. Average Mathematics Scores and Books in the Home: Grade 12 in 2005
Reported number Average mathematics
of books at home score
More than 100 163
26100 151
1125 136
010 126
Mathematics scores range from 0 to 300.
Table 7F. Average Science Scores and Books in the Home: Grade 12 in 2005
Reported number Average science
More than 100 161
26100 147
1125 132
010 122
Science scores range from 0 to 300.
Writing scores range from 0 to 300.
12th-giade scoies weie signicantly highei than foi households with 0-10 books. Tis
tendency applies to test-takeis in all foui subjects: math, science, civics, and histoiy.
At this point, we might ask whethei the ielationship between student test scoies
and numbei of books in the home is masking household income as a factoi in the
analysis. Te Depaitment of Educations NAEP Data Explorer tool, available on the
National Centei foi Education Statistics Web site, does not supply infoimation on
paients eainings, which we might have compaied with student test scoies. Yet the
tool does tiack anothei vaiiable closely ielated to socioeconomic status: paients level
of education.
Although test scoies iise successively with the paients education, the numbei of
books iemains a signicant piedictoi of bettei test scoies. Considei 12th-giade sci-
ence scoies: Students whose paients had high school diplomas and lived in homes
with moie than 100 books scoied 22 points highei than students whose paients weie
college giaduates and who iepoited 010 books in theii home.
Indeed, foi all the subjects iepoited in Table 7I, students of high school-educated
paients living in homes with moie than 100 books outscoied students with college-
educated paients and 010 books at home.
72
Students with high school-educated
paients and 26100 books in the home also outscoied students with college-edu-
cated paients and 010 books at home. Te ielationship between test scoies and the
availability of books may point to ieading and liteiaiy activity as highly visible func-
tions in the householdbehaviois that stiengthen the students own ieading habits.
Te close ielationship between student achievement and exposuie to books and
ieading has been quantied to some extent. But left untouched aie the less academ-
ically ielevant benets of ieading. Bettei test scoies aside, theie is a civic and social
impeiative, as subsequent pages will show.
72
Regiession analyses suppoit
this nding. Test scoies aie
ioughly 30 points highei foi stu-
dents with 100 oi moie books at
home than foi students with 010
books, iegaidless of theii paients
education.
Table 7G. Average Civics Scores and Books in the Home: Grade 12 in 2006
Reported number Average civics
More than 100 167
26100 150
1125 134
010 123
Civics scores range from 0 to 300.
Table 7H. Average U.S. History Scores and Books in the Home: Grade 12 in 2006
Reported number Average history
More than 100 305
26100 289
1125 275
010 265
History scores range from 0 to 500.
Reading the Numbers: Math and Reading Scores
In August 2007, the Department of Education released a report
comparing the U.S. educational system with those of the Group of Eight (G-8)
countries, which rank among the worlds most highly developed nations. The
report found a strong relationship between reading and math achievement in
seven of the G-8 countries: Canada, France, Germany, Italy, Japan, the Russian
Federation, and the United States. (The UK was omitted because of its low
response rate.)
Among 15-year-olds tested for reading and mathematics literacy in 2003, a low
score in either subject was usually accompanied by a low score in the other sub-
ject. For example, 82% of Americansthe highest percentage among G-8 coun-
trieswho scored at the lowest proficiency level in reading also scored at the
lowest level in math. Further, 62% of American 15-year-olds who scored at the
lowest level in math also obtained the lowest performance level in reading. For
details, see pp. 2728 of the report, Comparative Indicators of Education in the
United States and Other G-8 Countries: 2006, published by the Department of
Educations National Center for Education Statistics.
Table 7I. 12th-Grade Test Scores by Number of Books in the Home and by
Parents Education
(20052006)
U.S. History Civics
Test scores by Test scores by
parents highest level parents highest level
of education of education
High school College High school College
diploma graduate diploma graduate
Books in the home:
More than 100 290 309 152 172
26100 284 294 141 156
1125 270 281 133 140
010 264 275 121 131
Mathematics Science
Test scores by Test scores by
parents highest level parents highest level
of education of education
High school College High school College
diploma graduate diploma graduate
Books in the home:
More than 100 145 169 148 166
26100 143 157 141 152
1125 132 142 127 140
010 126 137 120 126
Data source: U.S. Department of Education, National Center for Education Statistics
Tobbe/zefa/Corbis
CHAPTER EIGHT
1. Employers nowrank reading and writing as top deciencies in newhires.
2. Good readers generally have more nancially rewarding jobs.
3. Less advanced readers report fewer opportunities for career growth.
BACKGROUND
cademic achievement tests aie useful in distinguishing between high and low
liteiacy levels in the geneial population, but they tell only half the stoiy. As
with many othei life skills, often of the aitistic oi intellectual kind, ieading
and wiiting uency yield intangible benets that shaipen the divide between those
who have the skills and those who do not. Some of those qualitative dieiences aie all
but immeasuiable, otheis can be iated by theii value in the job maiket.
Te impoitance of technological innovation to the national economy has ensuied
that math and science skills aie sought aftei and well-iemuneiated by many indus-
tiies. Incieasingly, howevei, think-tank oiganizations and expeit panels have also
cited the competitive value of advanced liteiacy in the Ameiican woikplace. Some
of theii conclusions diei on the type of liteiacy iequiied. In Closing the Expectations
Gap 2007, foi example, the nonpiot Achieve, Inc. iepoits:
Although [state] high school standaids and couises tend to emphasize liteiatuie,
most of the ieading students will encountei in college oi on the job is infoima-
tional in natuie (e.g., textbooks, manuals, aiticles, biiefs and essays).Most of
the wiiting students will do in college and woik is to infoimand peisuade, often
iequiiing students to use evidence to suppoit a position. Reseaich also is cited
as an impoitant skill foi college and woik. State standaids tend to give these
types of wiiting shoit shiift, emphasizing naiiative wiiting instead.
73
In contiast to this somewhat utilitaiian peispective, anothei nonpiot public pol-
icy gioup places cieativity on equal footing with tiaditional liteiacy skills. Te
National Centei on Education and the Economys 2007 iepoit, Tough Times or Tough
Choices, states:
Tis is a woild in which a veiy high level of piepaiation in ieading, wiiting,
speaking, mathematics, science, liteiatuie, histoiy, and the aits will be an indis-
pensable foundation foi eveiything that comes aftei foi most membeis of oui
woikfoice. It is a woild in which comfoit with ideas and abstiactions is the pass-
poit to a good job, in which cieativity and innovation aie the key to the good
life, in which high levels of educationa veiy dieient kind of education than
most of us have hadaie going to be the only secuiity theie is.
74
73
Achieve, Inc.: Closing the
Expectations Gap 2007, 9.
74
National Centei on Education
and the Economy, Tough Choices
or Tough Times: e Report on the
New Commission on the Skills of
the American Workforce, 2007,
xviii.
Src:io Tnvrr
Why Moie Tan Reading Is At Risk
A
Despite the iange of opinion on futuie caieei applications foi liteiacy, a ieview of
iecent employei suiveys is enough to attest that above-basic ieading and wiiting
skills aie in high demand. Depaitment of Education data pioling ieadeis at compe-
tent and decient liteiacy levels aie additionally helpful in identifying positive out-
comes associated with ieading and wiiting well. Key souices foi this chaptei aie:
Te Confeience Boaid, Coipoiate Voices foi Woiking Families, Paitneiship foi
21st Centuiy Skills, Society foi Human Resouice Management: Are ey
Really Ready to Work? Employers Perspectives on the Basic Knowledge and
Applied Skills of New Entrants to the 21st Century Workforce, 2006.
College Entiance Examination Boaid: Writing: A Ticket to WorkOr a Ticket
Out, A Survey of Business Leaders, 2004.
College Entiance Examination Boaid: Writing: A Powerful Message from State
Government, 2005.
U.S. Depaitment of Education, National Centei foi Education Statistics, Liter-
acy in Everyday Life: Results from the 2003 National Assessment of Adult Lit-
eracy, 2007.
Perceived Workforce Needs
Ameiican businesses and ieseaich and advocacy oiganizations posed a haid question
about Ameiicas new hiies in the title of a 2006 iepoit, Are ey Really Ready to
Work? In Apiil and May of that yeai, the gioup polled employeis foi peispectives
on the basic knowledge and applied skills of new entiants to the 21st-centuiy U.S.
woikfoice, a phiase captuied by the iepoits subtitle.
Te suivey sample was not nationally iepiesentative, and the iesponse iate was
low5, oi 431 companies with a combined woikfoice of 2 million U.S. employees.
Howevei, the sample was diveise: the highest peicentage of iespondents, 23,
belonged to the manufactuiing industiies, with anothei 21of employeis in govein-
ment, nonpiot, oi education sectois.
Business and piofessional seivices, on the one hand, and nancial and insuiance
seivices, on the othei, each claimed about 14of the iespondent pool, with some of
the iemaining peicentages distiibuted acioss the healthcaie (12), enteitainmenti
tiade (10), and eneigyiutilitiesiconstiuctionitianspoitation sectois (7). About
80 of companies weie small (less than 500 employees) oi mid-sized (5004,999),
with the iemaining iespondents each having 5,000 oi moie employees.
75
Despite the iange of employei-iespondents, a cleai majoiity viewed two basic skills
as veiy impoitant foi newwoikfoice entiants, iegaidless of education level. Tose
skills aie ieading and wiiting. See Table 8A.
Shown a list of nine basic skill aieasincluding math, science, and economics
90of employeis named wiiting in English as veiy impoitant foi the job success
of newhiies that had giaduated fioma foui-yeai college oi univeisity. Reading com-
piehension also ianked high among the basic skill needs foi this gioup, placing the
skill in the top thiee, just undei English Language skills.
We might expect college-educated employees to be valued moie highly foi theii
wiiting ability than employees with less education, if we can assume that the skill
was iened as pait of the college expeiience. But employeis expectations foi less-
educated woikeis, those who stopped at high school oi junioi college, also empha-
size the skill. Wiiting is the thiid most commonly selected skill deemed veiy
impoitant foi high school and junioi college giaduates alike. Among these hiies,
75
Foity-foui peicent of iespond-
ing employeis iecoided annual
ievenues undei $100 million,
while 29 had annual ievenues of
$1 billion and moie. Twenty-seven
peicent had ievenues fiom $100
million to $999,999,999. Peisonnel
with the job title of diiectoi
accounted foi just ovei half of
the suiveys iespondents. Othei
iespondents weie company vice-
piesidents, senioi VPs, manageisi
supeivisois, human iesouice spe-
cialists, and othei HR sta. (See
p. 61 of Are ey Really Ready to
Work?)
reading comprehension claimed the top spot, earning the most employer rat-
ings of very important.
In addition to asking employeis about basic skill needs, the suivey listed 11
applied skills such as ciitical thinking and pioblem solving, piofessionalism and
woik ethic, and asked which ones would be consideied veiy impoitant, impoitant,
oi not impoitant.
Reading compiehension was not among the applied skills fiom which employeis
could choose, but wiitten communications was. (Te suivey denes this skill as the
ability to wiite memos, letteis and complex technical iepoits cleaily and eectively.)
According to the majority of employers, written communications are essential
for newhires among two-year or four-year college graduates. See Table 8B. Moie
than 93of employeis iated the skill veiy impoitant foi job entiants with a college
education, while 72iated the skill veiy impoitant foi two-yeai oi technical college
giaduates. Foi two-yeai and foui-yeai college giaduates alike, wiitten communica-
tions ianks in the top ve applied skills valued most highly by employeis.
Foi new hiies whose education stopped at high school, employei expectations do
not appeai to value wiitten communications as gieatly as foi job entiants with highei
education. Yet just ovei half of iespondents (52) deemed the skill veiy impoitant
even foi this gioup of employees.
Top Deficiencies in New Hires
How do the skills of new hiies stack up with these peiceived job needs? Foi job
entiants of all levels of educational attainment, wiiting is faiing pooily. Among high
school giaduates, wiiting in English was the most commonly iepoited basic skill
Table 8A. Basic Skills Rated Very Important by Employers (Top Five)
Percentage of employers who rate the following basic skills as very important for:
High School Graduates Two-Year College Graduates Four-Year College Graduates
Reading Comprehension 63% Reading Comprehension 72% Writing in English 90%
English Language 62% English Language 71% English Language 88%
Writing in English 49% Writing in English 65% Reading Comprehension 87%
Mathematics 30% Mathematics 44% Mathematics 64%
Foreign Language 11% Science 21% Science 33%
Source: The Conference Board, Are They Really Ready to Work? (2006)
Table 8B. Applied Skills Rated Very Important by Employers (Top Five)
Percentage of employers who rate the following basic skills as very important for:
High School Graduates Two-Year College/Technical Four-Year College Graduates
School Graduates
Professionalism/Work Ethic 80% Professionalism/Work Ethic 83% Oral Communications 95%
Teamwork/Collaboration 75% Teamwork/Collaboration 83% Teamwork/Collaboration 94%
Oral Communications 70% Oral Communications 82% Professionalism/Work Ethic 94%
Ethics/Social Responsibility 63% Critical Thinking/Problem Solving 73% Written Communications 93%
Critical Thinking/Problem Solving 58% Written Communications 72% Critical Thinking/Problem Solving 92%
deciency iepoited by employeis (72). Wiitten communications, meanwhile, was
the most commonly iepoited applied skill deciency: 81%of employers reported
this problemwith high school graduates, as Table 8C notes.
Newhiies with two-yeai oi foui-yeai college backgiounds display a similai weak-
ness, accoiding to employeis, albeit at a lowei iate. Foity-six peicent of employeis
judged two-yeai college giaduates decient in wiiting in English, while 26 of
employeis iepoited this deciency in foui-yeai college giaduates. In both cases, the
basic skill ianked as the second gieatest deciency found in the new hiies.
Among applied skills, the job ielevance of liteiacy is even cleaiei. Foi both two-
yeai and foui-yeai giaduates, wiitten communications claimed the top spot in
applied skill deciencies cited by employeis. See Table 8D.
Befoie tuining to a sepaiate analysis of wiiting skill needs in the woikplace, we
should look moie closely at ieading compiehension, though not shown in the pie-
ceding table. As demonstiated in Table 8A, 63 of employeis ianked ieading com-
piehension veiy impoitant foi high school giaduates. Yet 38 considei most high
school giaduates decient in this basic skill.
Te suivey also ievealed that employeis in the manufactuiing sectoi weie moie
likely than othei employeis to iepoit ieading compiehension as a deciency in new
Table 8C. Skill Deficiencies in High School Graduates (Top Five)
Percentage of employers who rate high school graduates as deficient in basic and applied skills:
Basic Skills Applied Skills
Writing in English 72% Written Communications 81%
Foreign Languages 62% Leadership 73%
Mathematics 54% Professionalism/Work Ethic 70%
History/Geography 46% Critical Thinking/Problem Solving 70%
Government/Economics 46% Lifelong Learning/Self-Direction 58%
Table 8D. Skill Deficiencies in College Graduates (Top Five)
Percentage of employers who rate two-year and four-year college graduates as deficient in basic and
applied skills:
Two-Year College Graduates Four-Year College Graduates
Basic Skills Basic Skills
Foreign Languages 50% Foreign Languages 41%
Writing in English 46% Writing in English 26%
Government/Economics 32% Government/Economics 17%
History/Geography 26% History/Geography 17%
Mathematics 25% Science 13%
Applied Skills Applied Skills
Written Communications 47% Written Communications 28%
Leadership 43% Leadership 24%
Professionalism/Work Ethic 31% Professionalism/Work Ethic 19%
Lifelong Learning/Self-Direction 28% Creativity/Innovation 17%
Creativity/Innovation 28% Lifelong Learning/Self-Direction 14%
hiies. Foity-nine peicent of employeis in manufactuiing cited this basic skill as lack-
ing in job entiants, compaied with 36 of employeis in all othei industiies. Math
and science weie the two othei basic skills that manufactuieis found lacking in new
hiies moie often than othei industiies did.
Te impoitance of ieading compiehension to the manufactuiing sectoi is pio-
jected to giow, accoiding to a sepaiate suivey, the 2005 Skills Gap ReportASurvey
of the American Manufacturing Workforce, conducted by the National Association
of Manufactuieis and Deloitte Consulting LLP. In iesults fiomthe 800-employei sui-
vey, ieadingiwiitingicommunication skills ianked among the top thiee types of
skills that employeis will demand incieasingly. Moie than half of suiveyed employeis
said the need foi those skills would expand ovei the next thiee yeais.
76
Writing Skills in Demand
Chaptei Seven identied a coiielation between students ieading patteins and
National Assessment of Educational Piogiess (NAEP) wiiting test scoies. Anecdo-
tally, wiiting instiuctois have long witnessed the inspiiation, stylistic lessons, and
vocabulaiy giowth that the best student-wiiteis deiive fiom authois they iead on
theii own initiative.
In this iespect, ieading widely and ieading well can be a caieei investment. Two
iepoits, issued by the National Commission on Wiiting foi Ameiicas Families,
Schools, and Colleges, undeiscoie the peiceived demand of todays job maiket foi
advanced wiiting skills in newemployees. Te Commission was founded by the Col-
lege Entiance Examination Boaid in 2005, the yeai the boaid launched its new SAT
exam, which now includes a wiiting component.
Te ist iepoit, Writing: A Ticket to Workor a Ticket Out (2004), desciibed
iesults fioma mailed suivey with intensive phone call follow-up. Focused on wiiting
needs in the woikplace, the questionnaiie was distiibuted to 120 human iesouice
leadeis belonging to the Business Roundtable, a CEO membeiship oiganization.
Te suivey yielded a 53 iesponse iate. Like the suivey in Are ey Really Ready
to Work?, the wiiting data cannot be viewed as nationally iepiesentative, and a fui-
thei caveat applies. Unlike that pieviously discussed suivey, the wiiting suivey data
do not ieect a bioad iange of employment sectois.
Te National Commission on Wiiting notes, foi example, that manufactuieis aie
oveiiepiesented among Business Roundtable membeiship, and they accounted foi
69of suivey iespondents. An additional 15of iespondents belonged to the tians-
poitation oi utilities sectois, with anothei 7in nanceiinsuianceiieal estate, 6in
seivices, with the iemaindei in constiuction oi mining.
Some industiies with Business Roundtable membeiship weie not captuied by the
suivey: wholesale and ietail tiade and agiicultuie, foiestiy, and shing. Te Com-
mission also wains that the studys ndings cannot be applied to small business oi
goveinment employment. (Te aveiage employee base foi the 64 iesponding ims
was 54,503.)
Howdo human iesouice peisonnel in laige coipoiations iate the need foi wiiting
ability in the woikplace? Neaily 70 of iesponding ims said at least two-thiids of
theii employees have wiiting in theii job desciiptions.
77
Moie than half of iespon-
dents claimed they fiequently oi almost always considei wiiting ability in hiiing
decisions.
Accuiacy (95), claiity (75), and spelling, punctuation, and giammai (59) weie
the top thiee wiiting chaiacteiistics valued by employeis. Tey cited e-mail coiie-
76
National Manufactuieis Associ-
ation and Deloitte Consulting,
2005 Skills Gap ReportA Survey
of the American Manufacturing
Workforce, 2005, 8. See p. 25 foi
suivey methodology.
77
Unless noted otheiwise, the
employee data culled fiom both
National Commission on Wiiting
suiveys ielate to salaiied oi pio-
fessional employees, and not
houily oi cleiicalisuppoit sta.
spondence, piesentations and visuals, memos, and foimal iepoits as some of the
most fiequently iequiied communication foimats.
As foi woikeis wiiting ability, 34 of employeis iepoited that adequate wiiting
skills aie lacking in two-thiids oi moie of the cuiient woikfoice. Almost the same
peicentage36found wiiting skills lacking in at least two-thiids of incoming
employees. Te Commission concludes: Since up to one-thiid of the employees in
these blue-chip coipoiations do not possess adequate wiiting skills, wiiting decien-
cies may be even moie pionounced elsewheie in the bioadei piivate sectoi, paiticu-
laily among employees of small- and medium-sized businesses.
In 2005, the National Commission on Wiiting iepoited iesults fioma smallei but
moie iepiesentative employei suivey. Tis time, the Commission asked the National
Goveinois Association to poll state human iesouices diiectoisthiough an online
suiveyon wiiting skill needs of state goveinment employees. Foity-nine of 50 state
human iesouice diiectois iesponded.
Table 8E displays key iesults fiom the two wiiting suiveys side by side. On the
whole, state employeis appeai to value wiiting even moie highly than the laige com-
panies suiveyed. All of the state peisonnel diiectois said two-thiids oi moie of theii
employees have wiiting as a job iesponsibility. (Two-thiids of laige coipoiations said
the same foi theii employees.) Eighty peicent of state human iesouices diiectois fie-
quently oi almost always considei wiiting in hiiing new employees, compaied with
just 51 of laige companies.
When it comes to woikfoice wiiting ability, the coipoiate and state employei sui-
veys yield maikedly dieient conclusions: While 70 of state peisonnel diiectois
iepoited that two-thiids of cuiient and newemployees possess this skill, the shaie of
companies claiming this view is only half as gieat.
Costs of Remediation
Te compaiatively highei iate of satisfaction iecoided by state employeis foi theii
woikeis wiiting ability may ieect, at least in pait, a gieatei emphasis on tiaining.
Accoiding to the National Commission on Wiiting, two-thiids of state employei-
iespondents said they fiequently oi almost always piovide tiaining foi woikeis
whose wiiting abilities aie pooi, and less than 10of state employeis said they nevei
piovide such tiaining. Among laige coipoiations, just ovei 40fiequently oi almost
always piovide such tiaining foi employees who iequiie it.
Even if tiaining foi wiiting is emphasized moie stiongly by state employeis, the
ielative costs aie iepoitedly gieatei foi the piivate sectoi. State employeis iepoit pay-
Table 8E. Selected Results from National Commission on Writing Surveys:
A Comparison
Percentage who said: Corporate State
employers employers
Writing is a job responsibility for at least two-thirds of employees 67% 100%
Writing is frequently or almost always a factor in hiring decisions 51% 80%
Writing is frequently or almost always a factor in promotion decisions 50% 60%
The majority of new and current employees possess this skill 35% 70%
Some percentages are approximate.
Sources: College Entrance Examination Board, Writing: A Ticket to Work Or a Ticket Out (2004) and Writing:
A Powerful Message from State Government (2005)
ing, on aveiage, $425 pei employee foi wiiting tiaining, which the National Com-
mission on Wiiting estimates as costing state agencies $221 million annually.
Foi laige coipoiations, the aveiage spent on wiiting tiaining, acioss industiies, is
$950 pei employee, yielding a National Commission on Wiiting estimate of $3.1 bil-
lion pei yeai. Because of the iange of iepoited costs, howevei, the total estimates
must be viewed cautiously.
78
Prospects for Promotion
Accoiding to the National Commission on Wiiting, moie than 60of state employ-
eis take wiiting skills into account in making piomotion decisions, compaied with
almost half of laige coipoiate employeis. In industiy, manufactuieis aie moie likely
than the constiuction sectoi to base piomotion decisions on wiiting skills: 58 of
manufactuieis veisus one-thiid of constiuction companies.
Te Depaitment of Educations 2003 National Assessment of Adult Liteiacy
(NAAL) piovides anothei indiiect measuie of the link between liteiacy skills and job
advancement. In addition to gauging U.S. adults ieading ability, the NAAL culled
self-iepoited backgiound chaiacteiistics fiomthe test-takeis. Seveial of these back-
giound chaiacteiistics covei woikplace expeiience.
As shown by Table 8F, foi example, the majoiity of adults at the Below-Basic
ieading level (70) expiessed theii belief that low ieading skills had hindeied theii
job options. Tat peicentage dwindles in ielation to impioved ieading ability: 38of
Basic ieadeis said theii caieei giowth sueied, while only 16and 4of Inteimedi-
ate and Piocient ieadeis, iespectively, iepoited this expeiience.
Te NAAL also obtained self-iepoited data on salaiies of full-time employed adults
who took the test. Heie, too, advanced ieading skills coiielate with moie positive out-
comes, discounting othei factois. Te majoiity of Basic and Below-Basic ieadeis
62 and 77, iespectivelyiepoit eaining weekly incomes below $650, while the
majoiity of Piocient ieadeis fall within the $850-oi-moie-pei-week categoiy. Fifty-
seven peicent of Inteimediate ieadeis eain $650 oi moie pei week. See Table 8G.
What about the types of jobs that people of dieient liteiacy levels tend to have?
Accoiding to Table 8H, moie than 60 of Piocient ieadeis aie employed in the
management, business, and nancial oi piofessional and ielated sectois, wheie
aveiage weekly eainings aie typically the highest. Moieovei, piogiessively highei
iates of employment in those categoiies aie associated with highei ieading levels,
with 7of Below-Basic ieadeis, 18of Basic ieadeis, and 36of Inteimediate iead-
eis woiking in those sectois. Also notewoithy is that signicantly higher percent-
ages of readers at or above Basic are now employed in those sectors than in
1992.
79
78
Foi example, the Commission
iepoits that coipoiate wiiting
tiaining seivices iange fiom
online tutoiing piogiams costing
veiy little to full-scale wiiting
woikshops piiced in the thou-
sands. State agencies, by contiast,
iepoited estimates ianging fiom
$35 pei employee to $5,000 pei
employee. Te estimates include
houily oi cleiicalisuppoit sta,
and not just the piofessional oi
salaiied employees discussed
elsewheie in this section.
79
Te peicentage of Basic ieadeis
in management, business and the
nancial sectoi iose thiee points
fiom 5 in 1992, the peicentage of
ieadeis at that skill level who weie
employed in piofessional and
ielated elds giew two points
fiom 8 to 10ovei the same
peiiod.
Table 8F. Percentage of Adults Who Said Their Reading Skills Limited Their Job
Opportunities, by Reading Level in 2003
Prose literacy level Not at all A little Some A lot
Below Basic 30% 13% 22% 35%
Basic 62% 14% 15% 9%
Intermediate 85% 7% 6% 3%
Compaied with Piocient ieadeis, Below-Basic ieadeis do claim highei peicent-
ages of employment in ceitain sectoisconstiuction, pioduction, tianspoitation,
and seivice industiiesbut not in oce and administiative suppoit. Equal peicent-
ages of Below-Basic and Basic ieadeis, on the one hand, and Inteimediate and Pio-
cient ieadeis, on the othei, occupy the sales and ielated sectoi.
Fiomthese data, we cannot conclude that decient oi mediocie ieadeis univeisally
expeiience less job satisfaction oi less oppoitunity foi advancement than good iead-
eis. But when it comes to highei salaiies, self-iepoited condence in job success, and
employment in sectois with high-giowth potentialmanagement, business, and
piofessional caieeisieading skills aie in abundance.
Data fiomothei souices indicate that not only do employeis viewadvanced liteiacy
as ciitical to job success, but many Ameiicans iecognize theii ieading and wiiting
deciencies as potentially haimful to theii caieeis. Accoiding to suivey iesults
iepoited in 2005, one-thiid of high school giaduates who did not attend college iden-
tied the ability to iead and undeistand complicated mateiials as a gap in theii
piepaiation foi achieving goals in life. Even 29of college students iepoited this gap.
80
As foi wiiting ability, 35 of college students and 38 of high school giaduates
who did not attend college iepoited a gap between theii wiiting skills and the quality
of wiiting that is expected of them by futuie employeis, the suivey concluded.
An inteinational study, conducted by Statistics Canada with the Oiganization foi
80
Petei D. Hait AssociatesiPublic
Opinion Stiategies, Rising to
the Challenge: Are High School
Students Prepared for College and
Work? A Study of High School
Graduates, College Instructors,
and Employers, conducted foi
Achieve, Inc., 2005, 5.
Table 8G. Percentage of Full-Time Workers, by Weekly Earnings and Reading Level in 2003
Prose literacy level Less than $300- $500- $650- $850- $1,150- $1,450- $1,950
$300 $499 $649 $849 $1,149 $1,449 $1,949 or more
Below Basic 18% 41% 18% 12% 7% 3% 1% 2%
Basic 12% 31% 19% 16% 12% 5% 2% 4%
Intermediate 8% 19% 17% 18% 17% 8% 6% 8%
Proficient 4% 10% 11% 17% 20% 13% 13% 12%
Table 8H. Percentage of Adults in Each Occupational Group, by Reading Level in 2003
Prose literacy level Management, Professional Service Sales Office and
business, and financial and related and related administrative support
Below Basic 3% 4% 30% 8% 8%
Basic 8% 10% 24% 12% 14%
Intermediate 15% 21% 16% 12% 16%
Proficient 19% 42% 10% 7% 13%
Prose literacy level Farming, fishing, Construction Installation, Production Transportation
and forestry and extraction maintenance, repair and material moving
Below Basic 3% 15% 2% 15% 11%
Basic 1% 9% 4% 11% 9%
Intermediate 0% 5% 4% 7% 4%
Proficient 0% 3% 2% 3% 2%
Economic Coopeiation & Development, has attempted to quantify the distiibution
of decient ieadeis in the global woikfoice. By the studys estimates, 1920 of
Ameiicans ages 1665 aie mismatched foi theii jobs on the basis of theii ielatively
low scoies on liteiacy tests. Te studys authois explain:
Skill decits aie appaient in eveiy countiy, but the extent of the pioblemvaiies.
Appioximately 1030of the woikfoice can fall into this categoiy, depending
on the countiy. Some countiies have a compaiatively high skills decit.Pie-
sumably, a ceitain level of mismatch is expected in the laboi maiket but whethei
10, foi example, is noimal cannot be answeied with ceitainty. Highei iates,
howevei, aie likely to suggest a need foi adjustment, in paiticulai, the need foi
an incieased eoit to tiain peisons in those jobs.
81
81
Statistics Canada and the
Oiganisation foi Economic Co-
opeiation and Development,
Learning a Living: First Results of
the Adult Literacy and Life Skills
Survey, Copyiight OECD, 2005,
144.
Average Reading Scores: A Double Meaning?
In 2004, the Organisation for Economic Co-operation and Devel-
opment published Learning for Tomorrows World: First Results from the OECD
Programme for International Student Assessment. Known as PISA, the program is
a schools-based study the OECD conducted in 2000 and again in 2003 to assess
literacy skills and student achievement outcomes based on an internationally
accepted framework.
The 2003 assessment revealed that average American reading scores are truly
average. U.S. 15-year-olds barely placed in the top half of average reading scores
for 31 participating nations. Their scores lagged far behind those of readers in
such countries as Australia, Canada, Ireland, Korea, Finland, Sweden, and The
Netherlands. Especially in a global marketplace, with highly skilled and educated
workers increasingly in supply, such mediocrity may become untenable.
Table 8I. Average Reading Scores of 15-Year-Olds in Selected Countries: 2003
Country Score Rank Country Score Rank
Finland 543 1 Denmark 492 16
Korea 534 2 Iceland 492 16
Canada 528 3 Austria 491 17
Australia 525 4 Germany 491 17
New Zealand 522 5 Czech Republic 489 18
Ireland 515 6 Hungary 482 19
Sweden 514 7 Spain 481 20
Netherlands 513 8 Luxembourg 479 21
Belgium 507 9 Italy 476 22
Norway 500 10 Portugal 478 22
Switzerland 499 11 Greece 472 23
Japan 498 12 Slovak Republic 469 24
Poland 497 13 Turkey 441 25
France 496 14 Mexico 400 26
United States 495 15
Source: Learning for Tomorrows World: First Results from PISA 2003, Copyright OECD, 2004
CHAPTER NINE
5. Good readers play a crucial role in enriching our cultural and civic life.
6. Good readers make good citizens.
BACKGROUND
R
eadeis of poetiy may iecognize, in the second statement above, a faint allusion
to Robeit Fiosts poem Mending Wall. Te poem depicts two neighbois
mending a stone wall on a spiing day. Wintei has eioded the wall in paits,
leaving gaps that even two can pass abieast. As the neighbois toil on eithei side of
the wall, the speakei of the poem asks why a wall must divide theii piopeities in the
ist place.
Tis neighboi is moie open and playful and peihaps moie geneious than the othei,
who comes o as sti and piickly by compaiison. (In a chaiming metaphoi, the
speakei explains: He is all pine and I amapple-oichaid.) Te second neighboi justies
the wall, howevei, with a now-famous sentence: Good fences make good neighbois.
Te line iecuis at the end of the poem, but this time the meaning is iionic. By con-
tiast, no iiony is intended in the second statement heading this chaptei. Like the two
chaiacteis in Fiosts poem, advanced oi fiequent ieadeis and decient oi iiiegulai
ieadeis can be shown to exhibit unique sets of behaviois, and it is tempting to imag-
ine a wall dividing the two gioups.
Tis chaptei diaws fiom two publications, one an analysis conducted by the Aits
Endowment as pait of the 2002 Suivey of Public Paiticipation in the Aits, and the
othei a Depaitment of Education iepoit iefeienced in eailiei chapteis. Although
none of the data showcause and eect wheie ieading and ieadei tiaits aie conceined,
the two iepoits do highlight seveial shaied behavioi patteins linked with positive
individual, civic, and social outcomes.
National Endowment foi the Aits: e Arts and Civic Engagement: Involved in
Arts, Involved in Life, 2006.
National Centei foi Education Statistics: Literacy in Everyday Life: Results from
the 2003 National Assessment of Adult Literacy, 2007.
Literary Readers and Civic Engagement
Te 2002 Suivey of Public Paiticipation in the Aits (SPPA), designed by the National
Endowment foi the Aits and executed by the U.S. Census Buieau, is the most iecent
souice of nationally iepiesentative, statistically ieliable data on adult paiticipation in
aits events and activities. Foi the benet of aits oiganizations, aitists, academic
ieseaicheis, policy makeis, news media, and the geneial public, the 2002 data have
been iepoited in multiple NEA publications.
82
Tat yeai, the SPPAwas administeied as a supplement to the Census Cuiient Pop-
ulation Suivey. Te SPPA involved a sample of 17,135 adults (dened as 18 yeais of
age oi oldei), and the iesponse iate was 70. Because the suivey collected data not
only on aits paiticipation, but also on iespondent backgiound chaiacteiistics and a
iange of leisuie activities, the iesults peimit compaiisons of self-iepoited behavioi
82
Tey aie, in oidei of publication:
Suivey of Public Paiticipation in
the Aits (Reseaich Note #81),
Bonnie Nichols, Demogiaphic
Chaiacteiistics of Aits Atten-
dance (Reseaich Note #82), 2002
Survey of Public Participation in
the Arts (Reseaich Repoit #45),
Reading at Risk: A Survey of Liter-
ary Reading in America (Reseaich
Repoit #46), Bonnie Nichols, Aits
and Leisuie Activities: Evidence
fiom the 2002 Suivey of Public
Paiticipation in the Aits
(Reseaich Note, #89), and Bonnie
Nichols, Volunteeiing and Pei-
foiming Aits Attendance: Moie
Evidence fiom the SPPA
(Reseaich Note #94).
patteins acioss the geneial population. Fuithei, because the suivey has been con-
ducted ve times since 1982, tiend compaiisons aie available.
Te 2004 NEA iepoit Reading at Risk was based on an analysis of the 2002 SPPA
data and identied declines in liteiaiy ieading iates (the peicentage of adults who
had iead a novel, shoit stoiy, poem, oi play in the last 12 months) among Ameiicans
of both gendeis, all education levels, and most age gioups. Te iepoit also diew
attention to measuiable links between liteiaiy ieading and paiticipation in aits and
civic activity. Tose coiielations weie the focus of a sepaiate NEAieseaich biochuie,
e Arts and Civic Engagement (2006).
Table 9Ademonstiates the statistical stiength of those links, between liteiaiy iead-
ing and aits paiticipation and othei types of activities. Liteiaiy ieadeis aie well ovei
thiee times as likely as non-ieadeis to visit ait museums and attend plays oi musicals
oi classical oi jazz conceits. Although not shown heie, they aie even moie likely than
non-ieadeis to go out to the movies and listen to classical oi jazz iadio stations.
Fuithei, liteiaiy ieadeis aie signicantly moie likely than non-ieadeis to play
spoits oi attend amateui oi piofessional spoiting events. Tey do outdooi activities
(e.g., camping, hiking, and canoeing), exeicise at home oi in a gym, and cieate ait
thiough photogiaphs, paintings, oi wiitingsall at highei iates than Ameiicans who
do not iead ction, diama, oi poetiy.
Reading at Risk found that young adults (18- to 34-yeai-olds) in paiticulai weie
ieading liteiatuie at fai lowei iates than befoieat a 45 iate in 2002, the shaipest
decline (16 points) among all adults undei 65. Given the stiong coiielation between
liteiaiy ieading and othei positive individual and social behavioi, one would expect
to see declines in young adult paiticipation in those activities as well.
Tables 9B and 9Cbeai out this hypothesis. Liteiaiy ieading is among a host of cul-
tuial and civic activities that have expeiienced declines in young adult paiticipation
ovei a peiiod of two decades.
Readers Serve Their Communities
Volunteeiing is the most diiectly civic activity tiacked by the suivey. Heie the
iesults aie even moie stiiking than foi the othei leisuie activities done by liteiaiy
ieadeis. As Table 9D shows, liteiaiy ieadeis aie moie than twice as likely as non-
ieadeis to volunteei oi do chaiity woik.
Table 9A. Participation Rates for Literary Readers in 2002
Activity Literary readers Non-readers Gap between groups
Visit art museums 43% 12% -31 pp
Attend plays or musicals 36% 10% -26 pp
Attend jazz or classical concerts 29% 9% -20 pp
Create photographs, paintings, or writings 32% 10% -22 pp
Attend sporting events 44% 27% -17 pp
Play sports 38% 24% -14 pp
Exercise 72% 40% -32 pp
Do outdoor activities 41% 22% -19 pp
Te stiong link between liteiaiy ieading and volunteeiing was obseived in viitually
all demogiaphic gioups captuied by the SPPA: acioss all age gioups, education levels,
income gioups, and Census iegions. Moieovei, when contiolling foi education level,
gendei, age, and ethnicity, the study found that liteiaiy ieadeis weie thiee times as
likely as non-ieadeis to volunteei.
83
Finally, not only liteiaiy ieadeis but readers of
83
Bonnie Nichols, Volunteeiing
and Peifoiming Aits Attendance:
Moie Evidence fiom the SPPA
(Reseaich Note #94), 2006, 9.
Table 9C. Participation in Sports and Physical Activities (Ages 18-34)
Participation trends
1982 1992 2002 19822002 Rate of decline
Percentage who:
Attended a sporting event 62% 48% 43% -19 pp -31%
Played sports 60% 55% 43% -17 pp -28%
Did outdoor activities 50% 42% 38% -12 pp -24%
Exercised 67% 67% 60% -7 pp -10%
Table 9B. Literary Reading and Performing Arts Attendance (Ages 1834)
Participation trends
1982 1992 2002 19822002 Rate of decline
Percentage who read literature 61% 54% 45% -16 pp -26%
Percentage who attended a:
Jazz concert 16% 13% 11% -5 pp -31%
Classical concert 12% 10% 9% -3 pp -25%
Musical 19% 16% 15% -4 pp -21%
Ballet 4% 5% 3% -1 pp -25%
Table 9E. Percentage of Adults Who Volunteered, by Prose Literacy Level in 2003
Less than once Once a week Total who
a week or more volunteered
Basic 16% 15% 31%
Table 9D. Percentage of Adults Who Volunteered or Did Charity Work: 2002
43% 16% -27 pp
books in general aie moie likely than non-ieadeis to do chaiity woik. As Reading at
Risk demonstiatedsee pp. 67 of that eailiei iepoitvolunteeiing iates iose with
the numbei of books (liteiaiy oi nonction) iead ovei the past yeai.
Te Depaitment of Educations 2003 National Assessment of Adult Liteiacy
(NAAL) also contains statistics on volunteeiing. Figuies aie piesented not in ielation
to liteiaiy ieading, but by ieading skill level. Table 9E ieveals that not only do
advanced ieadeis, on aveiage, volunteei at a highei iate than Basic oi Below-Basic
ieadeisthey also volunteei moie fiequently.
Anothei measuiement of civic paiticipation is the peicentage of ieadeis who
choose to vote. Te assessment asked U.S. adults whethei they had voted in the most
iecent piesidential election, which had occuiied in 2000. As with the NAAL data on
Table 9F. Percentage of Adults Who Voted in the 2000 Presidential Election,
by 2003 Prose Literacy Level
Proficient 84%
Basic 62%
Below Basic 53%
Table 9G. Percentage of Adults Who Got Information About Current Events,
Public Affairs, and the Government from Various Media Sources, by Prose Literacy
Level: 2003
Source and literacy level None A little Some A lot
Newspapers
Basic 12% 23% 35% 30%
Below Basic 29% 25% 26% 20%
Magazines
Proficient 16% 37% 37% 10%
Basic 25% 30% 34% 12%
Below Basic 42% 23% 25% 9%
Books or Brochures
Basic 26% 30% 33% 11%
Below Basic 44% 23% 25% 8%
Internet
Proficient 16% 22% 31% 31%
Basic 53% 13% 17% 17%
Below Basic 77% 6% 9% 8%
volunteeiing, voting activity incieases in ielation to ieading skill level: fiom53and
62of Below-Basic and Basic ieadeis, iespectively, to 84of Piocient ieadeis. See
Table 9F.
Piocient ieadeis also exhibit gieatei cuiiosity about cuiient events, public aaiis,
and goveinment activity, as measuied by theii use of media to obtain this infoima-
tion. Table 9Gdisplays the peicentages of ieadeis at all skill levels who ieceived infoi-
mation on the afoiementioned subjects thiough thiee types of piint media and the
Inteinet.
Based on the links we alieady have obseived between voluntaiy ieading and iead-
ing piociency, the stiong ielationship between ieading skills and engagement with
piint media is less suipiising than anothei fact: even Internet usage for information
on current events, public aairs, and government rises in relation to reading
skillsfiom Below-Basic to Piocient.
Reading as an Act of Empathy
What accounts foi these dispaiities in civic engagement between decient and
advanced ieadeis? Foi some of the outcomes measuied in this chaptei, bioadei social
dynamics may be at woik.
Foi example, the lowei income levels often associated with less-skilled ieadeis may
iesult in less access to the Inteinet foi news puiposes, let alone othei types of com-
putei use. Oi the compaiatively lowei income level of the aveiage decient ieadei
may iequiie him oi hei to woik moie than one jobyielding fewei leisuie houis to
spend on volunteeiing and othei civic activities. Alteinatively, uniemaiked factois
unique to the household, neighboihood, oi health status of the individual may play a
iole.
In the absence of haid data explaining the reason foi incieased civic engagement
among liteiaiy and skilled ieadeis, we might considei the question moie abstiactly.
Te NEAs 2006 Arts and Civic Engagement ieseaich biochuie noted: By eveiy
measuie captuied by the Suivey of Public Paiticipation in the Aits, liteiaiy ieadeis
lead moie iobust lifestyles than non-ieadeis. Tese ndings contiadict commonly
held assumptions that ieadeis and aits paiticipants aie passive, isolated, oi self-
absoibed.
Anothei quotation, this one fiom the novelist, liteiaiy ciitic, and populai theolo-
gian C.S. Lewis (authoi of e Chronicles of Narnia), piesents a peisonal ieason foi
expecting ieadeis to identify moie closely with community than non-ieadeis:
Liteiaiy expeiience heals the wound, without undeimining the piivilege, of indi-
viduality. Teie aie mass emotions which heal the wound, but they destioy the
piivilege.But in ieading gieat liteiatuie I become a thousand men and yet
iemain myself.Heie, as in woiship, in love, in moial action, and in knowing, I
tianscend myself, and am nevei moie myself than when I do.
84
Good ieadeis, and not only liteiaiy ones, enjoy this piivilege of undeistanding and
appieciating the outlook of otheis while enlaiging theii own identity. Peihaps
because of this active empathy, they contiibute in measuiable ways to civic and social
impiovements.
Ultimately, ieading skills and eaily habits of leisuie ieading may come to occupy
the same ielationship to aitistic, cultuial, and civic piogiess as basic science skills
84
C.S. Lewis, An Experiment in
Criticism, Cambiidge Univeisity
Piess, 1995, 140141.
have had to technological bieakthioughs. Just as fundamental knowledge of math
and science has enabled piactical innovations in eveiyday life, so might young ieadeis
of today yield unfoieseen benets foi health policy, business, law, the social sciences,
aits and cultuie, jouinalism, and civic planning.
But why limit theii accomplishments to the humanities? In an eia of specialization,
the imaginative and analytical skills unlocked by ieading can fuel a biiskei dialogue
between the aits and sciences. As this iepoit has attempted to show, ieading often
and ieading well aie pieiequisites foi achievement in aieas fai beyond liteiatuie and
liteiacy alone.
Patterns of Disengagement: A Corollary
Despite positive findings about the civic engagement levels of
literary and advanced readers (see the preceding pages), the reverse trend for
deficient or non-readers is worthy of special attention. As the tables below indi-
cate, adults who are deficient readers are more likely than skilled readers to be
high school dropoutsand they are more likely to be absent from the workforce.
The NAAL recorded education and employment background statistics of test-
takers in 2003, and provided a comparison with 1992 levels. One-third of Basic
readers have a history of not completing high school, while one half of Below-
Basic readers share that history. Significantly, the percentage of Below-Basic
readers who are high school dropouts expanded by five points since 1992. (The
percentage of Proficient readers who dropped out of high school was only 1%.)
Measured on the basis of reading score, and not level of proficiency, the fol-
lowing table shows a gap of 55 points between the prose reading scores of 2003
high school graduates and high school dropouts. In addition to the gap, the aver-
age reading scores of both of those groups have decreased over time.
The nonprofit Editorial Projects in Education has reported that only 70% of all
high school students graduate on time with their peerssuggesting a high
dropout rate already for the general population. In this context, and given the
profound disadvantages facing high school dropouts later in life, the NAAL num-
bers for Below-Basic and Basic readers are especially grim.
These findings are significant because they stress the erosion of reading
scores for high school graduates and dropouts alike. The data also show that
Table 9H. Percentage of Adults at or Below Basic Prose Literacy Level
Who Did Not Complete High School: 1992, 2003
Prose literacy level
Below Basic Basic
1992 2003 Change 1992 2003 Change
45% 50% +5 pp 38% 33% -5 pp
Adults are defined as people 16 years of age and older living in households or prisons.
Table 9I. Average Prose Literacy Scores for Adult High School Graduates
and Those Who Did Not Complete High School: 1992, 2003
Prose literacy score
1992 2003 Change 9203
Gap between groups -52 -55
Adults are defined as people 16 years of age and older living in households or prisons.
dropouts retain far worse reading skills than graduates, which, as we have seen
in this report, places the former group at greater risk in a number of areas
throughout life.
Independent of what we have learned about employer preferences for good
readers, the following table shows a familiar pattern: greater societal risks
linked with poorer reading skills. Seventy-eight percent of Proficient readers
are employed, compared with far lower percentages of less-skilled readers.
Specifically, half of Americas Below-Basic readers and 38% of Basic readers are
classified as not in the labor forcea term that includes adults still in school
or keeping homecompared with only 18% of Proficient readers. An additional
6% of Basic readers and 5% of Below-Basic readers are unemployed or looking
for work, compared with 3% of Proficient readers. Although Chapter Eight pro-
vided some estimates on employer costs associated with remedial skills training,
there is no reliable estimate of the economic damage done to society by the lost
contributions of deficient readers.
Table 9J. Percentage Employed Full-Time or Part-Time, by 2003 Prose
Literacy Level
Proficient 78%
Basic 56%
Below Basic 45%
CONCLUSION
S
elf-iepoited data on individual behavioial patteins, combined with national
test scoies fiomthe Depaitment of Education and othei souices, suggest thiee
distinct tiends: a histoiical decline in voluntaiy ieading iates among teenageis
and young adults, a giadual woisening of ieading skills among oldei teens, and
declining piociency in adult ieadeis.
Te Depaitment of Educations extensive data on voluntaiy ieading patteins and
piose ieading scoies yield a fouith obseivation: fiequency of ieading foi pleasuie
coiielates stiongly with bettei test scoies in ieading and wiiting. Fiequent ieadeis
aie thus moie likely than infiequent oi non-ieadeis to demonstiate academic
achievement in those subjects.
Fiom the diveisity of data souices in this iepoit, othei themes emeige. Analyses
of voluntaiy ieading and ieading ability, and the social chaiacteiistics of advanced
and decient ieadeis, identify seveial disciepancies at a national level:
Less ieading foi pleasuie in late adolescence than in youngei age gioups
Declines in ieading test scoies among 17-yeai-olds and high school seniois in
contiast to youngei age gioups and lowei giade levels
Among high school seniois, a widei iift in the ieading scoies of advanced and
decient ieadeis
A male-female gap in ieading pioclivity and achievement levels
A shaip divide in the ieading skills of incaiceiated adults veisus non-piisoneis
Gieatei academic, piofessional, and civic benets associated with high levels of
leisuie ieading and ieading compiehension
Longitudinal studies aie needed to conimand monitoi the eects of these diei-
ences ovei time. Futuie ieseaich also could exploie factois such as income, ethnicity,
iegion, and iace, and howthey might altei the ielationship between voluntaiy iead-
ing, ieading test scoies, and othei outcomes. Ciitically, fuithei studies should weigh
the ielative eectiveness and costs and benets of piogiams to fostei lifelong ieading
and skills development. Foi instance, such ieseaich could tiace the eects of elec-
tionic media and scieen ieading on the development of ieadeis in eaily childhood.
Recent studies of Ameiican time-use and consumei expendituie patteins highlight
a seiies of choices luiking in the question to iead oi not to iead. Te futuie of iead-
ing iests on the daily decisions Ameiicans will continue to make when confionted
with an expanding menu of leisuie goods and activities. Te impoit of these national
ndings, howevei, is that ieading fiequently foi pleasuie is a behavioi to be cultivated
with the same zeal as academic achievement, nancial oi job peifoimance, and global
competitiveness.
I have sometimes dreamt, at least, that when the Day of Judgment dawns
and the great conquerors and lawyers and statesmen come to receive their rewards
their crowns, their laurels, their names carved indelibly upon imperishable
marblethe Almighty will turn to Peter and will say, not without a certain envy
when he sees us coming with our books under our arms, Look, these need
no reward. We have nothing to give them here. ey have loved reading.
Viiginia Woolf, How Should One Read a Book?
LIST OF CHARTS AND TABLES
27 Table 1A.
Peicentage of Adults Who Read a Book Not Requiied foi Woik oi School,
by Age Gioup: 1992, 2002
29 Table 1B.
Peicentage of Students Ages 9, 13, and 17, by Fiequency of Reading foi Fun:
1984, 1999, and 2004
30 Table 1C.
Peicentage of Students Ages 9, 13, and 17, by Pages Read Pei Day in School
and foi Homewoik: 1984, 1999, and 2004
31 Table 1D.
Peicentage Who Read Books foi Fun the Pievious Day: 2004
32 Table 1E.
Peicentage of High School Seniois Who Read foi Pleasuie, by Houis Pei Week:
1994, 2006
32 Chait 1F.
Peicentage of High School Seniois Who Read foi Pleasuie andioi Did Homewoik
foi 6 oi Moie Houis Pei Week: 19942006
34 Table 2A.
Fiist-Yeai College Students: Time Spent Reading foi Pleasuie: 2004, 2005
34 Table 2B.
College Seniois: Time Spent Reading foi Pleasuie: 2001, 2005
35 Table 2C.
Compaiison of Fiist-Yeai and Senioi College Student Reading Rates in 2007
36 Table 2D.
Compaiison of College-Piesciibed Reading Levels in 2007
36 Table 2E.
Compaiison of Time Spent on College Piepaiation in 2007
39 Table 3A.
Houis andioi Minutes Pei Day Spent on Leisuie and Spoits Activities,
by Age Gioup: 2006
39 Table 3B.
Houis andioi Minutes Pei Day Spent on Leisuie and Spoits Activities,
by Activity: 2006
39 Table 3C.
Houis andioi Minutes Pei Day Spent Watching TV oi Reading: 2006
40 Table 3D.
Peicentage of Daily Leisuie Time Spent Watching TV oi Reading: 2006
40 Table 3E.
Weekly Aveiage Houis andioi Minutes Spent on Vaiious Activities by
Ameiican Childien: 20022003
42 Table 3F.
Weekly Aveiage Houis andioi Minutes Spent Reading by Ameiican Childien, by
Age: 20022003
44 Table 3G.
Peicentage Using Othei Media While Reading: 20032004
44 Table 3H.
Peicentage of Time Spent Reading While Using Othei Media: 20032004
45 Table 3I.
Book-Reading Rates as Measuied by Multiple Suiveys
46 Table 4A.
Peicentage of Book Puichaseis, by Age: 1991, 2004
48 Chait 4B.
Unit Sales of Consumei Books: 2006
49 Table 4C.
Aveiage Annual Spending on Books, by Consumei Unit: 1985, 2005
50 Chait 4D.
Aveiage Annual Spending on Books, by Consumei Unit: 19852005
51 Chait 4E.
Aveiage Annual Spending on TViAudio Equipment Veisus Reading Mateiials
as a Shaie of All Enteitainment Spending, by Consumei Unit: 19952005
51 Table 4F.
Aveiage Annual Spending on TV and Audio Equipment Veisus Reading Mateiials
as a Shaie of All Enteitainment Spending, by Consumei Unit: 1995, 2005
56 Chait 5A.
Tiend in Aveiage Reading Scale Scoies foi Students Age 17: 19842004
57 Chait 5B.
Tiend in Aveiage Reading Scoies foi Students Ages 17 and 9: 19842004
58 Chait 5C.
58 Chait 5D.
60 Chait 5E.
Tiend in Aveiage Reading Scale Scoies foi 12th-Giadeis: 19922005
60 Table 5F.
Change in 12th-Giade Reading Scoies, by Peicentile: 1992 and 2005
61 Table 5G.
Peicentage of 4th-, 8th-, and 12th-Giadeis, by Reading Achievement Level:
1992, 2005
61 Table 5H.
Aveiage Reading Scoies of 4th-, 8th-, and 12th-Giadeis: 1992, 2005
62 Table 5I.
Aveiage 12th-Giade Reading Scoies by Gendei: 1992, 2005
64 Table 6A.
Peicentage of Adults at Each Piose Liteiacy Level: 1992 and 2003
64 Table 6B.
Aveiage Piose Liteiacy Scoies of Adults, by Highest Level of Educational
Attainment: 1992 and 2003
65 Table 6C.
Peicentage of Adults Piocient in Reading Piose, by Highest Level of Educational
Attainment (1992, 2003)
66 Table 6D.
Peicentage of Adults Piocient in Reading Piose, by Gendei (1992, 2003)
70 Chait 7A.
Aveiage Reading Scoies by Fiequency of Reading foi Fun: Age 17 in 2004
70 Chait 7B.
Aveiage Reading Scoies by Fiequency of Reading foi Fun: Giade 12 in 2005
71 Chait 7C.
Aveiage Reading Scoies by Fiequency of Reading Fiction Books oi Stoiies
Outside School: Giade 12 in 2005
72 Chait 7D.
Aveiage Wiiting Scoies by Fiequency of Reading foi Fun: Giade 12 in 2002
72 Table 7E.
Aveiage Mathematics Scoies and Books in the Home: Giade 12 in 2005
72 Table 7F.
Aveiage Science Scoies and Books in the Home: Giade 12 in 2006
73 Table 7G.
Aveiage Civics Scoies and Books in the Home: Giade 12 in 2006
73 Table 7H.
Aveiage U.S. Histoiy Scoies and Books in the Home: Giade 12 in 2006
74 Table 7I.
12th-Giade Test Scoies by Numbei of Books in the Home and by Paients
Education: 20052006
79 Table 8A.
Basic Skills Rated Veiy Impoitant by Employeis (Top Five)
79 Table 8B.
Applied Skills Rated Veiy Impoitant by Employeis (Top Five)
80 Table 8C.
Skill Deciencies in High School Giaduates (Top Five)
80 Table 8D.
Skill Deciencies in College Giaduates (Top Five)
82 Table 8E.
Selected Results fiom National Commission on Wiiting Suiveys: A Compaiison
83 Table 8F.
Peicentage of Adults Who Said Teii Reading Skills Limited Teii Job
Oppoitunities, by Reading Level in 2003
84 Table 8G.
Peicentage of Full-Time Woikeis, by Weekly Eainings and Reading Level in 2003
84 Table 8H.
Peicentage of Adults in Each Occupational Gioup, by Reading Level in 2003
85 Table 8I.
Aveiage Reading Scoies of 15-Yeai-Olds in Selected Countiies: 2003
87 Table 9A.
Paiticipation Rates foi Liteiaiy Readeis in 2002
88 Table 9B.
Liteiaiy Reading and Peifoiming Aits Attendance: Ages 1834 in 2002
88 Table 9C.
Paiticipation in Spoits and Physical Activities: Ages 1834 in 2002
88 Table 9D.
Peicentage of Adults Who Volunteeied oi Did Chaiity Woik: 2002
88 Table 9E.
Peicentage of Adults Who Volunteeied, by Piose Liteiacy Level in 2003
89 Table 9F.
Peicentage of Adults Who Voted in the 2000 Piesidential Election, by 2003
Piose Liteiacy Level
89 Table 9G.
Peicentage of Adults Who Got Infoimation About Cuiient Events,
Public Aaiis, and the Goveinment fiom Vaiious Media Souices, by Piose
Liteiacy Level: 2003
92 Table 9H.
Peicentage of Adults at oi Below Basic Piose Liteiacy Level Who Did Not
Complete High School: 1992, 2003
92 Table 9I.
Aveiage Piose Liteiacy Scoies foi Adult High School Giaduates and Tose Who
Did Not Complete High School: 1992, 2003
93 Table 9J.
Peicentage Employed Full-Time oi Pait-Time, by 2003 Piose Liteiacy Level

To Read

Caricato da

Informazioni sul documento

Descrizione originale:

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

To Read

Caricato da

Copyright:

Formati disponibili

National Endowment for the Arts

To Read or Not To Read

Potrebbero piacerti anche