The Pew Poll that Tom linked to yesterday is an interesting one.   It shows (to oversimplify somewhat) a 9% decrease in support for gun control since last year and an 8% drop in support for generally legal abortion (including an 11% drop since October).  While this is possible, I have to say the poll looks fishy to me.  Public attitudes about abortion are generally stable, and have been since the 70s.  It generallytakes a massive public discussion to move these numbers.

For example, in the mid-90s, there was a consistent decrease in the number of people who believed that abortion should be legal in all circumstances, and a concomitant increase in the number who believed abortion should be legal in most circumstances.  This coincided with the debate over the procedure known as either (depending on your political predilections) intact D & E or partial birth abortion.  Since a pretty small minority thought the procedure should generally remain legal, the number of people in the categorical camp on the abortion question declined somewhat.

In the absence of a similarly incendiary debate in the past few months, I find it hard to swallow that public support for legal abortion has declined to near-parity in a few months.  While support for gun control is a bit "noisier," the fact that the numbers on that question moved in parallel to the abortion question reinforces the notion that this movement isn't "real," but rather is the result of a skewed sample.  The drop of 24 points for support of legal abortion among moderate/liberal Republicans seems particularly unlikely.  More likely, Pew just wound up with a sample of self-described moderate/liberal Republicans who were particularly pro-life.

Every now an again, you'll draw a handful of only yellow M&Ms out of the bag, and every now and again you'll get a weird sample with your poll.  I think that's probably what's going on here.