Comments on Median earnings and selection biasTypePad2012-08-10T04:38:41Zjunkchartshttp://junkcharts.typepad.com/numbersruleyourworld/tag:typepad.com,2003:http://junkcharts.typepad.com/numbersruleyourworld/2012/08/median-earnings-and-selection-bias/comments/atom.xml/NB commented on 'Median earnings and selection bias'tag:typepad.com,2003:6a00d8341e992c53ef017c31aec8ed970b2012-09-06T12:03:27Z2012-09-07T03:19:49ZNB"The fact that the full-time median has stayed flat implies that it's the bottom that has fallen out. ... When...<p>"The fact that the full-time median has stayed flat implies that it's the bottom that has fallen out. <br />
...<br />
When the median number is changing as drastically as is portrayed here, we are looking at a crisis."</p>
<p>That is too big of a jump. Part time male workers have a declining median wage, yes, but there could be other elements at play. As more women enter ther workforce and households more commonly have multiple incomes (starting right about where the part time decline is in the 70s) you would expect to see declines in part time wages because competition just increased for the same jobs. Now with more families relying more on multiple incomes the individual earner's income doesn't need to be as high. </p>
<p>Is there still a problem, yes. But this doesn't necessarily imply an income crisis, there are a lot of other labor market features that tie in here. </p>Jeffrey commented on 'Median earnings and selection bias'tag:typepad.com,2003:6a00d8341e992c53ef016769406726970b2012-08-13T16:58:52Z2012-08-20T22:46:23ZJeffreyThe median is the midpoint of the data set. Because $25k and $0 are both on the same side of...<p>The median is the midpoint of the data set. Because $25k and $0 are both on the same side of the midpoint, movements from one of these levels to the other one will not affect the median. <br />
</p>jrkrideau commented on 'Median earnings and selection bias'tag:typepad.com,2003:6a00d8341e992c53ef0177440ff58d970d2012-08-11T13:50:28Z2012-08-13T03:27:54ZjrkrideauIf someone who's earning $25K (below median) becomes unemployed, this event has no impact on the median because earning $0...<blockquote> If someone who's earning $25K (below median) becomes unemployed, this event has no impact on the median because earning $0 also is a below-median number. </blockquote>
<p>Maybe I am not getting the context here but would this not actually raise the median? <br />
R code</p>
<p>df1 <- c(25,10,12,15,19,57,68,72,100,120,85)<br />
median(df1)</p>
<p>df2 <- df1[-1]<br />
median(df2)<br />
</p>Jonathan commented on 'Median earnings and selection bias'tag:typepad.com,2003:6a00d8341e992c53ef017617247750970c2012-08-10T13:52:02Z2012-08-13T03:27:54ZJonathan Kaiser check out some of the work by Rich Burkhauser on this. He has been discussing biases in this sort...<p>Kaiser check out some of the work by Rich Burkhauser on this. He has been discussing biases in this sort of measure for some time now. </p>