The US Department of Labor provides this great map of state minimum wage rates around the nation. Eighteen states comprising a majority of the nation’s population have minimum wage rates that are above the paltry $5.15, which is worth less and less every year.
Comments are closed.
I wonder what median pay rates look per state, and whether they correspond with minimum wage laws (and what the variance might be)?
That would be an interesting chart worthy of Kenton Ngo.
A chart would look a little like this.
Data from the Census Bureau and Wikipedia. A whole lot of states are preparing to hike their minimum wages soon — many on January 1.
As a quick-and-dirty calculation, I took the ratio of median income to minimum wage and multiplied the result by 100,000 (to avoid dealing with a really, really tiny number), giving me a number that indicates the spread between minimum wage and median income. The range turned out to be as low as 7.2 (the states with the highest median income and the lowest wages) and as high as 12.9 (the inverse.) At the top of the list is Arkansas, with a paltry $6.25 minimum wage, but a median income of $48k. Below that is West Virginia and D.C., both of which also have very low median incomes. Then Oregon, which has a $7.50 minimum wage and a $61k median income. At the very bottom of the scale is a tie between Colorado and good ol’ Virginia: we’re down at the federal minimum wage and median incomes of $71k apiece.
What would be more useful would be for somebody with a copy of SPSS to determine whether or not median incomes are a predictor of minimum wages, and which states are the outliers.
I’ll give it a shot, but I may be back with questions.
Nearly half the states (23) across a range of MI (45,867-79,339) stack up with a minimum wage at 5.15. So the data doesn’t follow much of a functional pattern – linear or otherwise – and regression isn’t all that informative. Leaving Kansas as an outlier, the data can be grouped in subpopulations that have significantly different average median incomes. One cluster (N=4) consists of those states with no minimum wage listed. Their average median income is 51,987. The next cluster is the group with the 5.15 minimum wage (N=23), and their average median income is 60,591. Finally the group with minimum wage strictly more than 5.15 (N=23) has average median income of 68,391.
I went ahead and used the Jan 2007 figure for Arizona for two reasons. First, it’s a conservative decision (as in depressing differences). And second, it gives a measure of what your after. But in the first pass the figures are existing numbers for minimum wage for the other states.
In any case, I repeated the process incorporating all of the Jan 2007 changes, and the analysis yields similar though weaker results. Kansas is again omitted. You get an R^2 of 18.87 on 46 states, lower R^2 (N decreased) in the right direction on the > 5.15 subgroup, but still the stacking is going on. Now the groups break out with:
No min wage (N=4): 51,987
Min wage = 5.15 (N=16): 59,919
Min wage > 5.15 (N=30): 67,072
Testing differences between means with a standard two-tailed t-test assuming unequal variance gives p for a two-tailed t-test is 0.0193. This is not as strong an effect between the analogous groups in the original data. The effect without the Jan 2007 increases was much stronger with p=0.0062.
So there’s a correlation, but it is not very direct (linear, polynomial, etc.) because the minimum wage variable is not strictly quantitative, but quasi-categorical in nature.
I hope this helps with what you are trying to do.
That is really interesting — thank you, Kathy! Your point on the quasi-categorical nature of the minimum wage variable is particularly valuable. You’ve given me flashbacks to my political science stats class, wherein I discovered that just because I can plug some numbers into SPSS and have some figures spat out does not mean that they actually mean anything. :)
FWIW, I ignored Kansas, since they’re superceded by federal law. They can set wages as low as they want, but it’s my understanding that nobody is affected it, since federal law takes precedence.
Good :) I’ve hit the wall in a nightmarish way on big data with SPSS, and I prefer SAS, but you can do all of this in a recent version of Excel if you install the data analysis add-in. Here’s a plot that is more informative, and if you right click on one of the data points in Excel you get many options, including a variety of regression line choices.
http://skatha.com/kath/minwage.png
You can use R http://www.r-project.org/ to do lot more.
Whoa.
You just blew my mind.
I’d never heard of R. I’ve only opened the front page and I’m already in love. It’s only Monday and it’s the best news of the week.
Free, open source, mature, and it runs on all three major platforms.
I’m downloading it now!