Thursday, December 31, 2009

Oregon, unemployment, and college

Once I get on a run, I tend to run with it.
Especially if it is something that I have data on, and can just cut and paste to different documents.
In the nation, there isn't much relation between college and unemployment, so lets look at the same data in Oregon.
(BTW, for those of you who don't know, I am an Oregonian. Mostly. Also, Oregon gives a good cross section of demographics.)Overall, there isn't a lot of direction in this diagram, although the four urban counties with most of the educated people and where much of the work goes on, are all grouped together in the middle.
So, again, another inconclusive result.

Oh, and also: 2010 is going to happen soon. It already did in London.
2010 means a number of things, including a CENSUS. However, the census results won't be out for two years, or so.

Monday, December 28, 2009

The same thing, with Oregon:

So, to make up for my terrible downturn in daily updating, and because I got interested,

I decided to look at the last plot I did, but for Oregon and its counties. Luckily, the census has growth figures by county, and the Department of Labor has a handy tool for looking at county level data

So the same graph for Oregon gives us:
There is a little bit more shape to this, but admittedly, not much. And again, although its not very clear, there is the same 5/6th pattern as we had last time: the only area where we don't see any points is high-growth, low-unemployment. Hood River and Benton Counties come close, though!
Another problem with this, as I have said for my Oregon diagrams before, is that bit all Oregon counties are equal in population. Especially noticeable in this, with Harney County, population 8,000, just sitting down there in the corner. I am tempted to do this graph with, for example, only the 10 or 15 most populous counties, and see what results I get from that.

Unemployment versus growth: once again, the easy explanation isn't the best one.

One piece of obvious conventional wisdom that I had been carrying around was that unemployment was higher in areas with high growth: that areas with high unemployment were areas with large influxes of population, and they therefore had high frictional unemployment, or had "oversold" themselves to potential workers. It seems like a good argument, and there are certainly a few data points to support it.
But you, my astute readers, know about "seems like a good argument" and "a few data points to support it". The actual scatter plot of the data is, as could be expected, scattered.
And, as is also often the case, there is a "three quarters" effect in here, although not a distinct one. Actually, It would be more a "five-sextet" effect. If we divide unemployment into high, medium and low, and growth into high and low...all of the five sextiles are occupied, except for "high-growth, low-unemployment". Utah, Texas and Colorado have high growth and medium unemployment, but there is nothing to the left of them.
Of course, in a normal economy, this graph might look different. So lets hope that we have a normal economy so I can find out! Also, I will have a job, and might not have time to scatterplot.
Incidentally, I found out this data several weeks ago, and just didn't bother to make a graph and post it. I actually have lots of stuff like that that I am sitting on!

Saturday, December 26, 2009

Content unrelated:

Now I am breaking down the sanctity of my scatterplot blog, just so I can post this image, and link from it for elsewhere.

But really, isn't it relevant?
Because aren't the data we mine out of the earth, the earth being random census documents, a secret to everyone? Like the money that rebellion-minded spear throwing dogs give to us?

Oregon is not Ohio. Neither is Washington.

After finding out that Ohio does indeed have a long pattern of following the nation's political trends, I decided to look at the same data for Oregon.
One thing to remember is that the electorate has been realigned many times since 1860 (which is as far back as I am going with these). Prohibition? Steel tarriffs? Vietnam? Female suffrage? The political and social and demographic issues that divide people have changed quite a bit over the years.
Which only makes it more important when a state does match up so closely with the nation. Whatever the political or social issue that divided the nation...Ohio somehow managed to feel pretty much the same way about it, since 1860.
I can't quite pin down a pattern to Oregon, though. Since 1980, it has been consistently more Democratic than the nation. Before that, it seemed to jump around, in a way that my knowledge of Oregon's demographics don't quite explain.
To wit:
Although, even with the fact that Oregon lines up less than Ohio does, there are still no major surprises here. While there are clusters of dots in the upper left and lower right quadrants, which represent not voting with the country, those dots are also pretty close to the origin, meaning that even though Oregon swung the other way (giggles), it didn't do so by a lot.
Along with that, I wanted to look at two states that would correlate with each other: Oregon and Washington. As expected,Oregon and Washington, going back to 1892, correlate pretty well. The major differences I think come from times when Washington was becoming industrialized, unionized and ethnicized before Oregon was, which gave it different demographics.

Thursday, December 24, 2009

Way to go Ohio...

One of the often quoted maxims of US politics is that the candidate that wins Ohio wins the nation. Of course, using modern technology, we can look at that a bit closer.
As can be seen, those numbers pretty much add up. They add up not just with the true/false test, but with the amount of the vote, as well. The exceptions are in a few landslide years, when Ohio doesn't always swing quite as wildly as the nation. But on the whole, it is true.
(This also explains the one exception to the rule: 1960, when Nixon won Ohio, but Kennedy won the nation. Nixon's victory in Ohio was small, as was Kennedy's victory overall).
According to the formula that shall not be named, there is quite a bit of correlation, and I bet that Ohio would indeed have a higher correlation than another other state, besides maybe Missouri.

I didn't need to snort all that GABA, anyway

I was e-Mailing with Dr. Stephen Wu, one of the authors of the "happiest states" reports, and I am happy to report that most of the spin put on the reports is due to the media, not to his research, which is much more modest in its claims.

Tuesday, December 22, 2009

Employment and hgihschool: once again, conventional wisdom reinstated

After finding out about the odd diamond shape that links college and unemployment, I decided to look at the same data and high school.

This gives us much more the shape we were thinking! States with lots of high school graduates have low unemployment! Unless they are Oregon and Michigan, of course.

Sunday, December 20, 2009

Education and unemployment:

Without doing the math for it (and you know I hate doing math), I would think the factorial combinations just of the things I have done so far would last me a good and long time. Election margins and U-3! Election margins and U-6! Wine consumption and election margins! Beer consumption and U-6!
Drowning in ideas, here. Which is perhaps why I have been actually posting less...I want to find things that are the most interesting, not just random shapes. (although random shapes are also good)
Anyway, here is something we want to know about, and that is also pretty interesting, shape wise:The bad news is: demographically, higher education levels (at the college level), don't seem to lead to higher employment. But then, neither do they lead to lower unemployment. In fact, we have an odd diamond shape: the states with the highest and lowest unemployment have about medium-range of college completion, while the states with the highest and lowest college completion have...a medium range of unemployment. There is also, somewhat meaningfully, a good amount of geographic/demographic clustering here, especially if you consider Colorado a Mid-Atlantic state (which you really should).

Obviously, like with all of my graphs, this bears more looking into.

Friday, December 18, 2009

Even with a broken monitor and a sprained finger, I can't let that type of foolishness go

You might think I am joking, but the "Happiest States" list annoyed me so much that I had to go and rail a line of GABA before I could deal with it.

I don't know if the original authors are as dumb as the media reports of it, but something tells me that they were not as critical of their own research as they should be. As for the media...

One problem with lists like these is that they are ranked cardinally, by order. Which is how sports teams are ranked. But, unlike making the playoffs, being ranked statistically is not so clean cut. When I get my hands on the actual data, I will plot that too, but as it is, I am just plotting the rankings, which might be very deceptive. There is a good chance that the separation on this list is by a small amount of degrees. This also comes up with lifespan measurements: a country may be ranked 30 places behind another country because people live three months shorter.

Anyway! I plotted the numbers against suicide rate. Suicide rate, is, of course, a bad way to measure people's general happiness. Suicide numbers are thankfully low, so even in a place where 99.9 of the people are very happy, a small subset might be suicidal. However, since suicide also probably correlates pretty well with suicide attempts, and major and minor depression, any measure of happiness that didn't somehow relate to it might be flawed.

So, I worked up this diagram:
On this diagram, the further to the left a state is, the more happy it is. The further up, the higher the suicide rate is. With that in mind, something should be noted: first, there is not much general trend at all. Second, what trend there is is towards the "unhappy" states having the lowest suicide rates. In fact, New York, the unhappiest state, has the lowest suicide rate. A big outlier like that should probably be a pretty big hint that there is something wrong with the data.

Of course, there are not many people reading this blog, so I am sure that BIG FANCY EAST COAST LIBERAL STATES=UNHAPPY BUT GOOD RELIGIOUS TRADITIONAL SOUTHERNERS=HAPPY will embed itself in at least some people's popular wisdom.

Oh, and the suicide statistics came from here:

I certainly have been slacking off, haven't I?

I have to admit that I have been slacking off.
One thing is, after doing so much research, I had lots of scatterplots, but I wanted to make them SPECIAL. No reason to just throw stuff up here.
Also, I sprained my finger
And, my monitor is broken.
But I will be back. Oh yes, I will be back, mostly because of this:

Which annoys me no end. In 2009, people ACTUALLY believe this? My god.
Of course, things like this will get repeated endlessly. And. The original data isn't even available. And. About three people read my blog of scatterplots. But. I must continue to FIGHT.

Saturday, December 12, 2009

U-3 and U-6, about what you would expect

As you might know, the "unemployment" numbers that are usually published are only one of the ways that unemployment is measured. It is technically called "U3", and covers people who are conventionally considered "unemployed". However, there are other numbers, ranging from U1 (Which is people who have not worked a single hour for pay in the last x months) to U6 (people who are underemployed).
I wondered how the different rates would match up, and thanks to the labor department I was able to find out:
As could be expected, the U-3 and the U-6 rates are very close to each other. Which makes sense, since U-6 by definition includes U-3. Since this diagram doesn't immediately tell us a lot, I took the U3/U6 ratio and plotted it against U3.
This diagram has some minimal good news: as a vague, general trend, as U3 goes up, the relative increase in U6 goes down. If the trend was flat, or even upwards, the underemployment ratio in Michigan could be over one-quarter.
But the trend isn't super important: as it is, it seems that within the limits we usually have, U3 and U6 go up pretty much up in tandem, regardless of what the number is.

Wednesday, December 9, 2009

Slight detour, Part II: the same thing, but different

So after yesterday's look at the urban/rural divide and college education, I did the obvious and did the same scatterplot for high school.
That is kind of a messy scatterplot! And not just because I made it at 2 AM! Within the expected confines, there seems to be a lot of variation in these numbers, more so than there is with college graduation rates. There is also some regional clumping, as could also be expected.
But, perhaps due to the lateness of the hour...I am not coming up with any magic bullets for this data.

Monday, December 7, 2009

A slight detour: rural and urban education

I became so besotted with the ERS and their gigantic stream of data, that I had to go slightly off the topic of farms, to find out about educational statistics, as they pertain to urban and rural America.
Did you know? "urban" and "rural" are hard statistics to operationalize. Which the ERS admits, they have an entire complex county-coding system. I can say, having lived in Montana and Vermont, both states that are considered "rural", that the word can mean very different things in different places.
But with those caveats aside, lets look at our scatterplot:As expected, urban areas have a higher percentage of people with bachelor's degrees (except in Massachusettes, where the rural population is almost non-existent, probably being the population of one resort community or something).
Also, Wyoming has perfect parity. The differences vary, from Virginia, with 5 urban degree holders for every 2 rural, up to states where the difference is almost unnoticable.
As could be expected, there seems to be some regional differences. New England and a chunk of Mountain/Plains states (Montana, Wyoming, Idaho, South Dakota) seem to have the smallest gaps, while the biggest gaps seem to be in Appalachia.

Sunday, December 6, 2009

Part II- hobby farms versus hobby farmers

I was thinking of a different way to phrase the question of how and why people farm...and a way that would work with the data presented by the USDA ERS. As I was drifting off to sleep, it occurred to me that it was pretty easy to operationalize "hobby farms" and "hobby farmers" with the data presented. One of the statistics is farms underneath 10,000 a year in SALES (that is gross, not net. 10K a year gross isn't a lot of money) and another is farmers who, as the saying goes "have kept their day job". The numbers of farms that produce less than a living amount, and the number of people who have to seek their living elsewhere, should more or less add up.
But you have been reading for a while, so lets see what really happens...

The basic view is, as is usually the case, more or less correct, but the trend doesn't jump out at me. Also, many of the states add up to much more than 100, meaning that there are lots of people whose primary income comes from a farm making less than 10K a year gross. I guess that would make these people more subsistence farmers than hobby farmers. At least in some cases; although it is hard to tell from the data presented.
Arizona is especially curious: I at first assumed that it was probably due to some loophole in zoning or tax laws in that state, but I later realized it might have to do with Native American subsistence farmers. Or, the data could have been a typo! Who knows!
It is also interesting to note that in very few states are most farmers primarily farmers, and in most states, most of the farms don't make much money.
"Further research is needed"

Saturday, December 5, 2009

Welcome back! For Laurel- Farm size versus ownership

I was gone trotting around Portland for two weeks, which meant that I had to leave you all scatter plot-less.
I hope this wasn't too sad.
So, today, I present the first of a series of scatterplots, requested by Laurel, centered around farming and the like. Specifically, she asked me about the link between non-corporate farming and farm output. I think. More or less.
So, thank you to , I have been able to start digging into this question. There will be more digging!
The first thing I wanted to look at was farm size versus private ownership. I would think that in the states where agriculture is a big business, farms would be larger and less privately owned.
I was, it seems, wrong. Farm size seems to be a lot more related to population density than anything else. Also, farm ownership seems to be pretty uniformly in the range of 80-90% private, across the board. Of course, some of those might be very small hobby farms. A median farm size would be an interesting thing to know.
Anyway, since this research wasn't very conclusive, I will play with more of the numbers in the coming days.