Thanks. That definitely shows the change, tho it doesn't look as dramatic as the other graph, which is why I suspect the other was used. It always peeves me when bad data representation is used to support a point, and it peeves me even more when it wasn't needed as is the case here.
Of course, there's still plenty of other things that could be done to massage the data, such as using a different definition of income, but at least then one is debating what should be measured, not how the resulting data is presented.
People looking for dramatic change like to show the actual dollars since the compounding effect of inflation will dominate the graph and the changes appear large. Once the result is controlled for inflation, as in my chart, more meaningful relationships can be seen. Personally I think my graph could make their case, but they would then have to dig deeper to say what had happened and what they would do to fix it.
As I noted on a similar thread on the Economics board, I think the data is quite telling. What it clearly shows is that income for unskilled and semiskilled work has lost value compared to skilled labor. I would like to see data from the first half of the twentieth century to see if the relationship was more stable during the era of heavy manufacturing and its need for semiskilled labor.