I love playing around with numbers and looking for patterns. Usually we are talking correlations, but under some correlations are causations. I haven't published much new playing with numbers for awhile, but in the past I played around with gender ratios around the world, and I have reviewed health outcome statistics and coverage vs. overall cost of health care vs. public funding of healthcare around the world.
Recently on the front page was the arresting article that the states with the lowest mortality rates are mostly "blue" and the states with the highest mortality rate are mostly "red." I decided to delve deeper into the numbers and look at the website cited in that article to find more information about different states in the US. Now, first thing I want to say is this is a "by eye" analysis. If someone paid me to do this I would do a far more extensive analysis, covering at least 10 years for each measurement, presenting graphs showing correlations and doing the necessary stats on these comparisons. But no one is paying me, so I am just doing a brief overview.
I also want to note that for almost everything I am looking at, the two most critical factors for good outcomes are education and lower poverty. I am not covering these two factors, though I would if someone was paying me! Every study I have looked at on the subject suggests that more funding to education and more funding towards alleviating poverty do more to reduce almost EVERY negative statistic you want to mention, which is why education, jobs, increasing minimum wage, etc. are critical for improving society. Ethnic issues and racism are also at play...another thing I would analyze in detail if I was paid for this!
But I decided to look at some other correlations. For each I start with an initial hypothesis but do not start with any idea whether my initial hypothesis will be supported by the numbers I am looking at. I did not have time to graph or do statistical analysis on the following. But some interesting things do come out of it. Some measurements I use only look at states, not territories like Puerto Rico, Virgin Islands, etc.
Some major numbers below.
REDUCING TEEN PREGNANCY:
First I want to look at the hypothesis that high teen pregnancy correlates with state policy that restricts abortions, access to contraceptives, and accurate sex education (as opposed to abstinence only education).
Of all developed nations, the US has one of the highest teen pregnancy rate, on average 3-fold higher than for the rest of the developed world. The trend has been VERY positive. Teen pregnancy rates have been going down across the board in the US for years now...but the discrepancies within the US are striking. The states that tend to be more conservative and more overtly religious have almost ALWAYS had and CONTINUE to have the higher teen pregnancy rates.
What I present below may not be the best comparison, but I look at teen pregnancy rate vs. NARAL ranking for each state. I look at the highest and lowest states (again if I was being paid I'd do a complete analysis) for teen pregnancy and compare with NARAL grades:
Teen births per 1000:
Below 25 teen births per 1000
NH: 15.7 (NARAL rating: B-)
MA: 17.2 (NARAL rating: B-)
VT: 17.9 (NARAL rating: A-)
CT: 18.7 (NARAL rating: A)
NJ: 20.1 (NARAL rating: A-)
ME: 21.3 (NARAL rating: A)
RI: 22.3 (NARAL rating: D+)
MN: 22.5 (NARAL rating: C+)
NY:22.7 (NARAL rating: A-)
(Almost all better than C, RI notable exception)
Above 50 teen births per 1000
Guam: 60.1
MS: 55.0 (NARAL rating: F)
NM: 53.0 (NARAL rating: A-)
AR:: 52.5 (NARAL rating: F)
TX: 52.2 (NARAL rating: F)
PR: 51.4
VI: 50.5
OK: 50.4 (NARAL rating: F)
(Almost all F, NM notable exception)
BOTTOM LINE: States that are most opposed to NARAL policies tend to have higher teen pregnancy rates. Conservative policies on sexuality tend to correlate with higher teen pregnancy (would be better if could get similar numbers for access to contraception and for accurate sex ed classes vs. abstinence only).
DEATH:
Second I want to look at the hypothesis that states with worse insurance coverage have a higher death rate. This is in some ways an upfront measurement of the value of health insurance, but it also is over simplified since bother values I am trying to compare will be influenced by other factors such as poverty, racism, and education. But let's take a look.
This is roughly what the dKos article I cite at the beginning, that inspired this diary, was covering. The message of that article was that the states with the lowest mortality rates were bluer than the states with the highest mortality rates. I wanted more info. So I decided to look at insurance coverage. Is there a correlation between being covered by insurance and lower death rate. The answer is "yes" but it is reasonably subtle in the numbers. My first thought was just to look at the % uninsured in 2012 (the latest number) and try to correlate with death rate...there is some correlation, where states with a lower % uninsured had a lower death rate, but it was not easy by eye to split the best from the worst states based on one year. So I looked over 14 years and compared how many years each state had fewer than 10, 15 or 20% uninsured and tried to correlate with death rate. The correlation kind of balances around 15% uninsured...states that have a hard time maintaining that number tend to also have some years where the uninsured rate gets upwards of 20 % and they also tend to have a higher death rate. States that consistently maintain less than 15% uninsured and even reach 10% uninsured tend to have the lowest death rates.
Death rate: deaths per 100,000
Below 700
HI: 589.6 (% uninsured 2012: 7.7; 14/14 years below 10)
CA: 646.7 (% uninsured 2012: 17.9; 14/14 years below 20)
CT: 652.9 (% uninsured 2012: 8.1; 10/14 years below 10; 14/14 years below 15)
MN: 661.5 (% uninsured 2012: 8.3; 14/14 years below 10)
Virgin Islands: 663.2
NY: 665.5 (% uninsured 2012: 11.3; 13/14 years below 15; 14/14 years below 20)
MA: 675.0 (% uninsured 2012: 4.1; 13/14 years below 10; 14/14 years below 15)
CO: 682.7 (% uninsured 2012: 13.7; 7/14 years below 15; 14/14 years below 20)
NH: 690.4 (% uninsured 2012: 12.0; 9/14 years below 10; 14/14 years below 15)
NJ: 691.1 (% uninsured 2012: 14.0; 12/14 years below 15; 14/14 years below 20)
WA: 692.3 (% uninsured 2012: 13.6; 14/14 years below 15)
AZ: 693.1 (% uninsured 2012: 18.0; 13/14 years below 20)
Almost always below 20% (only AZ goes above 20% and even there only 1 year). Most are consistently below 15% uninsured (CA a notable exception), about half are below 10% for at least half that time.
Above 850
MS: 962.0 (% uninsured 2012: 15.3; 1/14 years below 15; 12/14 years below 20)
AL: 939.7 (% uninsured 2012: 14.8; 11/14 years below 15; 14/14 years below 20)
WV: 933.6 (% uninsured 2012: 14.6; 11/14 years below 15; 14/14 years below 20)
American Samoa:932.9
OK: 915.5 (% uninsured 2012: 17.2; 1/14 years below 15; 14/14 years below 20)
KY: 915.0 (% uninsured 2012: 15.7; 10/14 years below 15; 14/14 years below 20)
LA: 903.8 (% uninsured 2012: 18.3; 1/14 years below 15; 11/14 years below 20)
AR: 892.7 (% uninsured 2012: 18.4; 2/14 years below 15; 14/14 years below 20)
TN: 890.8 (% uninsured 2012: 13.9; 2/14 years below 10; 13/14 years below 15; 14/14 years below 20)
Northern Marinas: 863.3
SC: 854.8 (% uninsured 2012: 14.3; 7/14 years below 15; 13/14 years below 20)
Almost never below 10% uninsured, all go above 15% uninsured at least once, about half go above 15% more than half of the time. Occasionally some go above 20%.
Bottom line: there is indeed a correlation between % insured and lower death rate, but it seems there are also other things going on (I suspect education and poverty may be impinging...possibly ethnic issues and bigotry as well.)
DEATH BY FIREARMS:
I believe the 2nd Amendment should be treated just like the 1st Amendment and all other parts of the Constitution. But I also want to know the numbers. Here I want to look at the hypothesis that states that have more lenient firearm laws have more deaths due to firearms than states with strict firearm laws. I also want to look for correlations between firearm laws and crime rates. Again I am not taking the time to do as thorough an analysis as I would if this was my job, but I think the results are interesting.
Deaths due to firearms per 100,000
Below 7
HI: 3.2 (Violent crime rate per 100,00: 239.2; Property crime rate per 100,00: 3,075.2; Guns and Ammo ranking: 46)
MA: 4.1 (Violent crime rate per 100,00: 405.5; Property crime rate per 100,00: 2,153.0; Guns and Ammo ranking: 48)
RI: 4.6 (Violent crime rate per 100,00: 252.4; Property crime rate per 100,00: 2,572.3; Guns and Ammo ranking: 42)
NY: 5.1 (Violent crime rate per 100,00: 406.8; Property crime rate per 100,00: 1,922.0; Guns and Ammo ranking: 50)
NJ: 5.2 (Violent crime rate per 100,00: 290.2; Property crime rate per 100,00: 2,047.3; Guns and Ammo ranking: 49)
CT: 5.9 (Violent crime rate per 100,00: 283.0; Property crime rate per 100,00: 2,140.0; Guns and Ammo ranking: 45)
IA: 6.8 (Violent crime rate per 100,00: 263.9; Property crime rate per 100,00: 2,271.8; Guns and Ammo ranking: 38)
MN: 6.8 (Violent crime rate per 100,00: 230.9; Property crime rate per 100,00: 2,568.3; Guns and Ammo ranking: 39)
Above 15
AK: 20.4 (Violent crime rate per 100,00: 603.2; Property crime rate per 100,00: 2,739.4; Guns and Ammo ranking: 3)
LA: 19.2 (Violent crime rate per 100,00: 496.9; Property crime rate per 100,00: 3,540.6; Guns and Ammo ranking: 17)
AL: 16.2 (Violent crime rate per 100,00: 449.9; Property crime rate per 100,00: 3,502.2; Guns and Ammo ranking: 7)
MS: 16.1 (Violent crime rate per 100,00: 260.8; Property crime rate per 100,00: 2,811.0; Guns and Ammo ranking: 18)
WY: 15.6 (Violent crime rate per 100,00: 201.4; Property crime rate per 100,00: 2,293.8; Guns and Ammo ranking: 6)
MT: 15.4 (Violent crime rate per 100,00: 272.2; Property crime rate per 100,00: 2,583.7; Guns and Ammo ranking: 11)
Not a major correlation between crime rates and firearm deaths, though the states with the top three firearm deaths per year ALSO have higher violent crime rates than any of the low firearm death states. But AL vs. MS and HI vs MA show the correlation may be misleading. (NOTE this study showing a correlation between firearm ownership and homicide rates for the 1990's: https://www.ncbi.nlm.nih.gov/... ; also see: http://www.sciencedaily.com/... )
Good correlation between firearm friendliness (based on Guns and Ammo rankings) and firearm deaths. ALL of the states with the highest firearm deaths are considered very gun friendly by Guns and Ammo. ALL of the states with the lowest in firearm deaths are less firearms friendly than half of the states in the union, and include the LEAST firearm friendly states. But this is not the whole story since, for example, VT is number 2 firearm friendly and isn't high in firearm deaths. But it is one of the strongest correlations I could find.
I remain convinced that poverty (as addressed by things like minimum wage and safety nets) and education (as addressed by large increases in education funding) are the keys to most of these issues. But I also think better sex ed and more lenient laws on abortion and contraception availability DO help reduce teen pregnancy, more health insurance coverage DOES help reduce death rates, and the MOST lenient gun laws correlate with higher firearm death rates but NOT with reduced crime rates.
Other factors are involved, but we can't ignore these numbers.