The numbers in your figure come from a recent Georgetown University study (instead of the BLS). Below, is the methodology from their report:
Methodology
Data from the American Community Survey for the years 2009 and 2010 were pooled to provide a larger sample size for the estimates. The unemployment rates were then computed for each of the three groups by dividing the total unemployed with the total employed and unemployed. The earnings used are median earnings in 2010 dollars rounded to the nearest $1,000. The three groups are: recent college graduates (those between ages 22 and 26 with bachelor degrees), experienced college graduates (those between ages 30 and 54), and graduate degree holders (those with master’s degrees or higher and are between 30 and 54). Median earnings are based on those who worked more than 35 hours a week and at least 50 weeks a year. All calculations use the survey weights provided by the Census Bureau.
I still have no clue how many former students (i.e. the sample size) were surveyed and from which schools. Personally, I think the listed unemployment rates for "recent college graduates" are too low (all across the board) and the earnings numbers are too high. Okay, I know the figure you provided doesn't show earnings rates, however, I did look at them from the report I mentioned above.
The construction industry has been in the sh*tter for years throughout the country, so, I don't believe that too many people need a survey to be told that employers are not hiring students with Architecture majors.
Maybe someone here can tell me what is the purpose of conducting unemployment surveys. If you look at the numbers on the chart you could be led to believe that it doesn't really matter what type of degree you earn (if you are simply looking for a job and don't care about earnings) because besides the Architecture and Arts major all of the numbers are clustered close together.
