With young people, employment numbers are a little tricky. I think it’s better to look at >=25yos with college degrees, for which FRED provides a data series from the BLS where the values range from ~1.4% to ~5.4% and currently is 3.3%. Depending on how you interpret the survey responses (only explicit “unemployed” or non-responses too?), LWers in the same group (>=25yo, with a bachelors or higher degree) seem to have ~5-7%:
R> survey2013 <- read.csv("http://dl.dropboxusercontent.com/u/182368464/lwsurvey/2013.csv″, header=TRUE)
R> age ← survey2013[survey2013$Age>=25,]
R> degree ← function (x) { x!=” ” & x!=”2 year degree” & x!=”High school” & x!=”None” }
R> bachelors ← age[sapply(age$Degree, degree),]
R> length(is.na(bachelors$WorkStatus) | bachelors$WorkStatus==” ”)
[1] 700
R> sum(!is.na(bachelors$WorkStatus) & bachelors$WorkStatus!=” ”)
[1] 691
R> levels(bachelors$WorkStatus)
[1] ” ” “Academics (on the teaching side)”
[3] “For-profit work” “Government work”
[5] “Independently wealthy” “Non-profit work”
[7] “Self-employed” “Student”
[9] “Unemployed”
R> sum(bachelors$WorkStatus==”Unemployed”,na.rm=TRUE)
[1] 38
R> sum(bachelors$WorkStatus==”Unemployed” | is.na(bachelors$WorkStatus) | bachelors$WorkStatus==” ”)
[1] 47
R> c(38/691, 47/700)
[1] 0.05499 0.06714
R> binom.test(38,691, p=0.033)
Exact binomial test
data: 38 and 691
number of successes = 38, number of trials = 691, p-value = 0.002649
alternative hypothesis: true probability of success is not equal to 0.033
95 percent confidence interval:
0.03921 0.07470
R> binom.test(47,700, p=0.033)
Exact binomial test
data: 47 and 700
number of successes = 47, number of trials = 700, p-value = 6.492e-06
alternative hypothesis: true probability of success is not equal to 0.033
95 percent confidence interval:
0.04975 0.08829
(I wonder if I’m using the best BLS data-series, though; they record a lot of data and there can be subtleties that outsiders don’t appreciate.)
With young people, employment numbers are a little tricky. I think it’s better to look at >=25yos with college degrees, for which FRED provides a data series from the BLS where the values range from ~1.4% to ~5.4% and currently is 3.3%. Depending on how you interpret the survey responses (only explicit “unemployed” or non-responses too?), LWers in the same group (>=25yo, with a bachelors or higher degree) seem to have ~5-7%:
(I wonder if I’m using the best BLS data-series, though; they record a lot of data and there can be subtleties that outsiders don’t appreciate.)