In my column last week, I explained that Los Angeles Times reporter and former Laura Bush press secretary Andrew Malcolm took a few liberties with polling data in order to make things look bleak for President Obama.

Today, Malcolm is back at it. Malcolm headlined his post about a new Washington Post poll “9 of 10 Americans worry about Obama's spending deficits: Poll.” Later, he wrote “Currently, 90% of Americans are worried to some degree about the exploding federal spending deficit.”

Actually, he made that number up. The Post poll found 87 percent are concerned about the deficit. 87 is not 90. Granted, it's awfully close, and not substantively different. But reporters can't just go around changing poll results to fit their whims.

Far more substantively: Malcolm's headline misstates the poll results. Malcolm claims the poll found that 90 percent of Americans “worry about Obama's spending deficits.” Actually, the poll didn't attribute the deficits to Obama (or to spending, for that matter -- deficits are not only about spending, they're about tax cuts, too.)

Malcolm's implication that Americans overwhelmingly worry about Obama's handling of the deficit (reinforced by his later assertion that “particular unhappiness focused on his handling of ... the federal deficit”) is flatly contradicted by the poll's actual findings, in which the public is split down the middle on that question: 48 percent of Americans approve of the president's handling of the deficit, and 48 percent disapprove.

It is, however, consistent with Malcolm's assertion last week that “As the months roll by, the results, added together, the clock is running out on Obama's ability to blame the last administration for all ills; the sense of his ownership of the nation's problems appears to be growing in the American mind.” That assertion, by the way, was directly contradicted by the polls Malcolm was citing, which is probably why he didn't include any actual poll numbers to buttress the assertion.

But that's not all.

Malcolm stipulates that “Obama's personal popularity remains high,” though he doesn't give an actual number. Or even a made-up number that is a few points lower than the actual number. In any case, he is again misrepresenting the poll, which did not ask “personal popularity” (whether people have a favorable or unfavorable impression of Obama.) It asked whether they approve of the job he is doing. Here's the actual question: “Do you approve or disapprove of the way Barack Obama is handling his job as president?” The poll found 65 percent approve.

But Malcolm is trying to convince readers that though Obama is personally popular, his policies and job performance are unpopular. So he falsely claims the poll found 90 percent concern about Obama's deficits (rather than 87 percent concern about the deficit.) And he changes the approval question, too, portraying it as a question about “personal popularity” rather than job approval.

The moral of the story: If Andrew Malcolm writes about poll results, you should assume that he's either lying, or he doesn't know what he's talking about. Or both.