From the Department of Labor:
SEASONALLY ADJUSTED DATA
In the week ending February 11, the advance figure for seasonally adjusted initial claims was 348,000, a decrease of 13,000 from the previous week’s revised figure of 361,000. The 4-week moving average was 365,250, a decrease of 1,750 from the previous week’s revised average of 367,000.
UNADJUSTED DATA
The advance number of actual initial claims under state programs, unadjusted, totaled 361,928 in the week ending February 11, a decrease of 39,328 from the previous week. There were 424,400 initial claims in the comparable week in 2011.
Last year’s seasonally adjusted figure for the comparable week was 420,000.
Seasonal factor adjustments accounted for the entire reported improvement:
Last year’s factor (confirmed by going to the interactive table here) was 1.010 (424,400 above divided by 420,000) This year’s is 1.040 (361,928 divided by 348,000). If last year’s factor had been applied to this year’s data, SA claims would have been 358,000 (361,928 divided by 1.010), the same as last week’s 358,000 figure before this week’s report upwardly revised it to 361,000.
As I’ve noted similarly in other circumstances, in the case of  both 2012 and 2011 we’re looking at work weeks in early February unaffected by holidays. The only reason the seasonal adjustment factors are so different is that the past almost four years of recession-driven and weak “recover”-driven data entering the seasonal adjustment machine have been so volatile and atypical.
Though I probably won’t do it because of time constraints, it almost seems worth it to run a shadow report on this year’s raw claims using last year’s seasonal factors to see if DOL is consistently using higher ones to reduce reported seasonally adjusted numbers, and to see what their longer-term impact is.
After filtering for seasonal factor chances, and though the doubts about how many people are still out there to get laid off persist — because, as explained last week, covered employment has barely moved off of its low last year, and is still down about 5% from where its 2007 peak — today’s result was decent, but not the spectacular improvement the press will probably portray.
_________________________________
UPDATE: An interesting observation about how the nation’s good luck in avoiding serious winter weather has affected things –
The numbers, he says, indicate that fewer construction workers and others lost their jobs because the weather was good. In fact, he says, only 206,000 people were unable to work because of the weather last month, compared with an average of 425,000 the previous five Januarys, according to the BLS household survey.
“You have to look at these numbers with a grain of salt,†says Mr. Brown. “If this hiring continues in February, March, and April, that would be really something.â€
The luck we’re having this winter is being at least offset, possibly moreso, by the horrid, deadly winter in Eastern Europe (“European cold snap has claimed more than 650 lives”).
UPDATE 2: Well, here’s a partial look at the seasonal adjustment issue, namely a graphic of the raw factors for last year vs. this year thus far accompanied by an evaluation of their impact —

Only one difference inflated claims, and it was the one affected by the New Year’s holiday, which received relatively little attention because everyone understands that week’s volatility. Three weeks’ year-over year claims were significant (enough to deflated reported seasonally adjusted claims by 10,000 or more), and the other two had a negligible impact.
As they say, “developing …”