[AAACE-NLA]How Well Can Adults Read?

[Thomas Sticht]

Research Note				            2 May 2003

How Well Can U. S. Adults Read?
Government-Centered vs. Learner-Centered Estimates
Tom Sticht

The 1992 National Assessment of Adult Literacy (NALS) tested a
representative sample of adults 16 years of age and older on three tests
of literacy: Prose, Document, and Quantitative. Scores on each test were
divided into five levels of literacy, with Level 1 being the lowest level
and Levels 4 and 5 the highest levels. Results on the three types of tests
were very similar so only the data for the Prose test are used here.

Using the Prose test, some 21 percent (40+ million) of U. S. adults were
placed in Level 1 and 27 percent (50+ million) in Level 2. These two lower
levels included almost half  (47-48 percent) of the adults in the U. S.
and this was the basis for a Press Release of 8 September 1993 from the U.
S. Department of Education with the headline LITERACY LEVELS DEFICIENT FOR
90 MILLION U. S. ADULTS.

Interestingly, the Press Release went on to say , "
most of these adults
describe themselves as being able to read or write English "well" or "very
well." The Press Release goes on to quote Secretary of Education Richard
W. Riley  making the extraordinary statement: "It paints a picture of a
society in which the vast majority of Americans do not know that they do
not have the skills they need to earn a living in our increasingly
technological society and international marketplace." In short, this
statement about adults’ insights into the adequacy of their skills simply
dismissed the adults’ judgements about their skills in favour of the
results of the standardized tests made by the government sponsored testing
experts.

Over the  decade from 1992 up to the present, numerous reports using the 
data for adults on the Prose, Document and Quantitative tests have
appeared and the NALS methodology was also used in developing the
International Adult Literacy Survey (IALS) which was eventually
administered in over 20 nations.  For the U. S. a subset of the NALS data
for 16 to 65 year olds were used in the IALS.  Like the NALS, the IALS
used the three literacy tests as their primary means of literacy
assessment, and they also used self-ratings of how well adults thought
they read. And again, as with the NALS, the IALS research has lead to many
reports about the percentages of adults in various nations with
"deficient" literacy skills using the test data, but almost no attention
has been given to the self-rating data presented in the technical reports.

The present paper brings together data from analyses of the respondents to
the Prose test in various reports from the NALS and IALS in a summary
manner regarding how well adults in the U.S. think they read. These data
reveal some interesting and important differences among adults in the U.S.
in their beliefs about their reading skills. The data are summarized below
in a number of questions and answers.

Q1: How many  adults 16 years of age and above in the U. S. think they
read English "well" or "very well?"
A1: 93 percent thought they read "well" or "very well’ while 7 percent
(13.4 million) thought their reading was deficient. This differs greatly
from the government’s estimate of 90 million adults with deficient reading
skills based on its standardized tests.

Q2: How many adults 16-65 years of age think their reading skills meet
their DAILY needs for reading "moderately well" "good," or "excellently"?
A2: 94 percent "moderately," "good," or "excellently;"  87 percent "good"
or "excellently."

:Q3: How many adults 16-65 years of age think their reading skills meet
their main JOB needs "moderately well," "good," or "excellently"?
A3: 94 percent "moderately," "good," or "excellently;"  87 percent "good"
or "excellently."

Q4: How many adults 16-65 years of age think their reading skills "are not
at all limiting"  "somewhat limiting"  or "greatly limiting" of their job
opportunities?
A4: 87 percent "not at all limiting,"  9 percent "somewhat limiting,"  4
percent "greatly limiting."

Q5: How many adults aged 16-59 who were White, Black, or Hispanic thought
they read English "well" or "very well?"
A5: Whites, 98 percent, Blacks, 94 percent, Hispanics, 68 percent.

Q6: How many adults aged 60 years or age or higher who were White, Black,
or Hispanic thought they read English "well" or "very well?"
A6: Whites, 95, Blacks, 83, Hispanics, 54 percent.

Q7: How many prison inmates thought they read English "well" or "very well."
A7: 88 percent.

Q8: How many adults age 16 years or higher said they got "a lot"  or
"some/none" help with completing literacy forms of some sort?
A8: 12 percent "a lot,"  88 percent "some/none."

Clearly, the government’s test-based estimates of adult reading/literacy
skills differ considerably from the adult population’s experience-based
estimates of their reading/literacy skills. This raises the question of
how each type of information might be best used in establishing the scale
of need for resources for adult literacy education.

Of course, it might be argued that neither type of information is the best
information for establishing the scale of need for adult literacy
education and that instead we ought to focus on the numbers of adults who
present themselves for educational services each year and argue for
funding needed to provide adequate services for those who present
themselves for education.

I see these as important issues for adult literacy providers and adult
learners to deal with in advocating for adult literacy education funding.
I wonder what others think of the general issue of how we should go about
determining the scale of need for adult literacy education services. Is
this an important issue? How should it be approached?

Finally, if some 9 out of 10  adults think they read "well" or "very
well,"  and that their reading skills meet their daily needs and their
needs at work, and that their reading skills do not limit their job
opportunities, then what does this imply for policy for reading
instruction in the K-12 system, which is where most adults acquire most of
their reading skills. Does this suggest that there is a need for massive
reforms in the teaching of reading in the K-12 system, ehther they are
evidence-based or not?

Thomas G. Sticht
International Consultant in Adult Education
2062 Valley View Blvd.
El Cajon, CA 92019-2059
Tel/fax: (619) 444-9133
Email: tsticht at aznet.net