also from the same article: agreement with question

[victor steinbok]

This is not particularly uncommon IMO and is more understandable in the
context of public surveys, where every items is called a question, even if
it may actually contain an affirmative statement.

http://goo.gl/qPI3D

> Although two-thirds of Americans say that Muslims are not trying to
> establish Shariah law in the U.S., "Over the last 8 months agreement with
> this question has increased by 7 points, from 23 percent in February 2011 to
> 30 percent today."



A more clear phrasing would have been "agreement in this question" or "on
this question", but I can understand the reluctance to go that route.

The way these surveys are conducted, the respondent is asked:

Do you agree or disagree with the following statement?
> Muslims are trying to establish Shariah [law] in the United States.


Or, perhaps, they are asked,

On a scale of 1 to 5, 1 being strongly disagree, 2--somewhat disagree,
> 3--neither agree nor disagree, 4--somewhat agree, and 5--strongly agree,
> please rate the following statement:

Muslims are trying to establish Shariah [law] in the United States.


Either way, the agreement is with the statement.

"Question" and "equation" were my favorite pet peeves in education. All test
items are "questions" even when they are not at all questions, and all
"questions" that ask for some kind of computation or symbol manipulation are
"equations" even if they involve no equality of any kind (in particular,
when someone asks for a formula, they often ask for an "equation"--some are,
some aren't; but 5x+3 is not an "equation" by any stretch of imagination). I
still correct people about "equation", but I see "question" as a part of
accepted practice and psychometric jargon. But here, the issue is not so
much the "question", as "agreement with the question".

VS-)

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org