[Reading-hall-of-fame] Key PIAAC Finding: Is Reading Not a Major Problem In Adult Literacy?

[tsticht at znet.com]

10/17/2013

Key PIAAC Finding: Is Reading Not a Major Problem In Adult Literacy?

Tom Sticht        International Consultant in Adult Education

The Program for the Assessment of Adult Competencies (PIAAC) used a set of
tasks involving information processing skills including reading in the
assessment of two types of literacy tasks: prose literacy and document
literacy.  To better understand how an adult’s reading skills, independent
of the other complex information processing involved in performing the
literacy tasks, related to the literacy performance, the PIAAC also studied
what were called the components of reading.

Earlier, the International Adult Literacy Survey (IALS), which formed the
technical methodology for the  Program for the Assessment of Adult
Competencies (PIAAC) literacy assessments, also assessed the components of
reading thought to underlie the performance of the reading aspects of the
complex literacy tasks in the IALS. In the IALS study the components of
reading were defined as word recognition, ( recognition of both pseudo and
real words), vocabulary knowledge, spelling, and digit span (a traditional
measure of working memory capacity).  Interestingly, in the IALS reading
components study, the digit span measure of working memory correlated more
highly with the literacy tasks than did the word recognition or vocabulary
knowledge measures. Surprisingly, the spelling component of reading had the
highest correlation with the IALS measures of literacy.

The PIAAC used a different set of reading components from those used in the
earlier IALS research.  The components were described and a sample given as
follows:

Print vocabulary:  Items testing print vocabulary consist of a picture of an
object and four printed words, one of which refers to the pictured object.
Respondents are asked to circle the word that matches the picture. 
Example: The examinee sees a picture of an ear and has to circle one of the
following four words: ear  egg  lip  jar


Sentence processing: The sentence-processing items require the respondent to
assess whether a sentence makes sense in terms of the properties of the real
world or the internal logic of the sentence. The respondent reads the
sentence and circles YES if the sentence makes sense or NO if the sentence
does not make sense. Examples: Here are some sample sentences:
Three girls ate the song. YES NO
The man drove the green car. YES NO
The lightest balloon floated in the bright sky. YES NO
A comfortable pillow is soft and rocky. YES NO


Passage comprehension:  In items assessing passage comprehension,
respondents are asked to read a passage in which they are required, at
certain points, to select the word from the two alternatives provided that
makes the most sense. Here is an example: “To the editor: Yesterday, it was
announced that the cost of riding the bus will increase. The price will go
up by twenty percent starting next: wife / month. As someone who rides the
bus every day, I am upset by this: foot /increase. I understand that the
cost of: gasoline / student has risen.”

The results were reported as the average proportion of reading component
items answered correctly, by literacy proficiency level. For the United
States, average proportion correct for print vocabulary increased from 90.4
for those in Below Level 1 (the lowest level of literacy skill), to 96.2 for
Level 1 adults,  98.6 for Level 2,  99.8 for Level 3, and 99.8 for adults in
Levels 4/5 combined (the highest levels of literacy skill).


For sentence processing, US adults scored in the same levels as above,
respectively: 72.4,  83.0,  93.0, 96.5,  and 97.5.


For passage comprehension, US adults scored in these same levels,
respectively:  82.2,  90.3,  96.7,  99.5, and  99.9.


Taking the average as representative of the abilities of adults in each
level of literacy skill suggests that “reading” , defined as the measures
of reading components used in the PIAAC, is not a major problem for most of
the adults in the US nor any of the other OECD nations involved in the
PIAAC. Even for those in the lowest level of literacy skill, those Below
Level 1, the average percentage correct for all the OECD nations on Passage
Comprehension was 87.9 percent, which is well above the 80 percent correct
criteria that is frequently considered as “mastery”  by instructional
developers.

This raises the question, if reading is not the major problem for those
performing on the PIAAC literacy tests, then what is it that is creating
the distribution of scores from the lowest to the highest levels of
literacy?  The identification of different sets of components of reading in
the IALS and the PIAAC surveys suggests confusion about what exactly the
components of reading really are and how they relate to the literacy tasks
used in the PIAAC. For instance, why didn’t the PIAAC assess digit span and
spelling as components of reading like the IALS study did? This apparent
uncertainty about just what the components of reading are raises an
important question. If it is not clear what the components of reading
really are, and if reading was not a major problem for adults taking the
PIAAC, how do adult literacy educators know what to teach to help adults
improve their literacy skills as they are assessed in these international
assessments?

tsticht at aznet.net