I have to admit that it is difficult for me to keep up with who said
what, and whose conclusions are whose. Yet the results don't seem that
puzzling to me. For example, these two statements: 

1) There was no overall decline in the average score for prose literacy

2) The score for prose literacy dropped at every level of education

To me, the first statement precludes one from suggesting that prose
literacy has declined. In fact, given the reports of the scores of
African Americans (increases in all three), Asians and White (a mix of
increases and unchanged), it would seem that things are not so bad. The
reduced average scores of Hispanic adults is troubling, but I think the
combination of a more inclusive study and changes in demographics might
have more to do with it than any generalized decline in the nation's
literacy (defined by this instrument). 

I also think that more than likely the second statement does not provide
evidence of a reduction in standards or of a reduced impact of higher
education. To measure the impact of higher education by this instrument,
you would have to track individuals' progress through the school system.
That is, what score did you have before you went to school and what did
you have after? This is also true at the societal level, if one is
trying to measure the impact of college spending on general literacy
levels. Again, since the study did not involve identifying what literacy
level was required by colleges, inferences about changes in the nature
of those requirements would not be supported by the data. 

I think that framing it in terms of standards might confuse the analysis
of the decline at the different educational levels. Imagine that there
was a group of people in 1992 who had no college experience despite
having literacy skills that could possible provide entry. They didn't
have the high scores in literacy typically associated with college, but
were more on the line between going and not going. Now that more members
of this group are in college (the report notes more people falling in
that category), it would seem to do two things. First, it would reduce
the literacy levels of "those without college experience" because the
higher scorers from that category have been removed. A good thing, since
that means they are getting a chance to go to college. Second, the
average scores of those with college experience might be reduced because
this newly entering group is not beginning college with the same scores
of those that had been there in the past. Again, since there is no
preset score that indicates that one has met the standards for entering
college, I don't think you can automatically say that standards for
admission have declined. 

The increased number of people who might show up as "high-education with
lower scores" (and therefore reduce the average score of the
high-education group) would only be a bad thing if once they were in
college their literacy scores did not go up, but this report cannot
address that because it is a not a longitudinal study - it is snapshot.
It is possible that although their scores are lower compared to past
averages, the college experience has increased their personal score
quite a bit. Of course it is also possible that it didn't, but we don't
know. Because of this ambiguity, I don't think one can jump to
conclusions about lower standards or a lowered impact of higher
education based on the results of this study. 

Erik Jacobson