While I don’t disagree with the fact that if you eliminate certain demographics from the equation you’ll end up with a very different answer, I question the merits of that answer. Specifically with regard to stripping out all part-time workers based on the notion that *some *(what percentage?) of those workers are not relying on their incomes to meet basic needs such as shelter, food and clothing. By eliminating this demographic, you also eliminate those who are subsisting only by working multiple part-time jobs or basically just trying to stay off (or stretch) unemployment benefits.
To counter your anecdotal data of your mother, I already indicated that I personally am supporting my family (barely) on a part-time job combined with my spouse’s unemployment benefits. More than a few of the people I work with are working multiple part-time jobs to cover their living expenses and only a portion of those individuals are college students or retirees. None of the part-timers I work with are there because they don’t *need *the income which is something we share in common with the full-time management staff. The company I work for employs over ten thousand people, most of whom work in retail outlets of which the majority of those employees are part-time. I like my job a lot and if I were offered full-time I’d take it, but as I previously noted, it’s unavailable to me in my current position. However, even if I worked full-time at my current rate of pay, I would still earn less than $26,000/yr.
How does eliminating me and my co-workers from the statistic provide a more accurate picture of state of wage income in America? And why should as little as 156 hours/year be the sole justification for including me in your revised statistics?
Possibly that isn’t the best way to do it, the point being that there are many populations in this data - other populations missing from this data - and this data doesn’t end up saying much at all about the underemployed because it isn’t good data for that. We don’t know if the median has decreased since 1999 because the economy sucks (probably a big factor) or because baby boomers have started to wind down their careers for lower impact jobs (probably a small factor) or because efforts toward immigration reform have put thousands of migrant workers “on the books” who were previously under the table (I don’t know if that’s a factor) off this information. If you start chopping up the data, and combining it with other sources of data, you might be able to draw conclusions, but right now, its frankly sort of a floating data point.