The point is not who has the misperceptions, or why...

The point is how "misperceptions" are defined by the authors in the first place.

Sorry, Tom... that's your point, and I'm not buying. The point of the study is precisely "who has the misperceptions, [and] why." The conclusion is: the degree of misperception is dependent upon both the source of information and the attention paid to it.

Here are the definitions you choose to attribute to "positions" of NPR/PBS (neither organization was involved in the survey, btw - see the page immediately preceding the report proper)...

Page 1:

More striking, in PIPA/KN polls conducted after the war -- in May, July, and August-September -- approximately half of the respondents expressed the belief that the US has actually found evidence in Iraq that Saddam was working closely with al-Qaeda. While administration figures have talked about a purported meeting in Prague between an al-Qaeda member and an Iraqi official, this does not constitute evidence that Saddam was working closely with al-Qaeda and, in any case, this purported meeting had been discredited by the US intelligence community during the period of these polls.

[emphasis added]

Page 4:

Perhaps the most striking misperception is the belief that, not only were there links between Saddam Hussein and al-Qaeda, but that actual evidence has been found. PIPA/KN asked in May, July, and August-September: “Is it your impression that the US has or has not found clear evidence in Iraq that Saddam Hussein was working closely with the al-Qaeda terrorist organization?” In every case, large percentages (45-52%) said that such evidence has been found.

Page 4-5:

Before the war overwhelming majorities believed Iraq had weapons of mass destruction. Though it now appears this belief may have been incorrect, it does not seem appropriate to call this a misperception because it was so widespread at the time, even within the intelligence community.

However, a striking misperception occurred after the war, when the US failed to find any weapons of mass destruction or even any solid evidence of a WMD program.

As an aside, note the graph on page 3 showing the public view of Iraq/Al Qaeda connections when the respondents were asked more detailed questions. The chart shows a more realistic perception than the 70 percent quoted as believing in a direct link between Iraq and 9/11. The middle two positions (65 percent total) are reasonable ones, and the 30 percent position is factually confirmed.

Note that the first two categories of response in that graph - neither of which is supported by fact - adds to 57%. As to the "70%", this is from the text immediately below that graph (page 3, right-hand column):

A widely noted August Washington Post poll asked, “How likely is it that Saddam Hussein was personally involved in the September 11 terrorist attacks? Would you say that it is very likely, somewhat likely, not very likely, or not at all likely?” Thirty-two percent said it was very likely and another 37% said it was somewhat likely. This has led to a widely-repeated statement that 7 in 10 Americans believe that Saddam Hussein was personally involved in September 11. However, it should be noted that when respondents say that something is likely -- especially those who just say that it is somewhat likely -- it does not mean they have come to the conclusion that it is the case.

[emphasis added]

Concerning the "factually confirmed" 30%, I refer you back to the quote from Page 1 which appears above.