Thursday, May 18, 2017

How is Washington DC an outlier? Let's count the ways. (Repost from Data Driven Journalism)

My latest post on Data Driven Journalism is up an reprinted here.
 
In my last post, I reported that Washington, DC had an extremely high rate of 30.83 hate groups per million residents in 2016 relative to the other 50 states (the national rate was 2.84 groups per million).  DC also had an exceptionally low percent of the vote for Donald Trump in 2016, at just 4.1%.  For these reasons, and other characteristics which make DC fundamentally different from the other 50 states, I had to exclude it from a correlational analysis between hate group concentration and Trump’s percent of the vote.  For this post, I will look at other ways in which DC is an outlier.

According to the most recent Small Area Income and Poverty Estimates (SAIPE) from 2015, DC ranks third in median household income at $70,848 behind Maryland and Alaska. Yet, the same SAIPE estimate also ranks DC eighth for the percent of the population in poverty, at 17.4%.  This indicates a large gap between the rich and poor.  The high rate of poverty is reflected in DC’s low life expectancy at 76.53 years, ranking 43rd compared with the overall US average of 78.86 years. Similarly, DC’s infant mortality ranked eleventh in the country, at 7 deaths per 1,000 live births compared to the US rate of 5.9 deaths per live births.  Newly released estimates from the Census Bureau for 2015 show DC has the second lowest rate of those without health insurance at 4.3% behind Massachusetts. These income and health statistics suggest that DC deviates from the national rates, but not that it is an extreme outlier – with one exception.

The statistics on crime suggest that DC is an extreme outlier.  DC had a violent crime rate of 1,244.4 offenses per 100,000 residents in 2014.  This is almost twice as large as the next highest state, being Alaska with a rate of 635.8 offenses per 100,000 residents, and more than three times as large as the US rate of 365.5 offenses per 100,000 residents.  In 2014, it had the highest murder rate of any other state at 15.9 offenses per 100,000 residents. 



Image: Paul Ricci.
Last fall, the FBI’s Uniform Crime Report released the number of hate crime incidents in 2015 for each state.  Adjusting their numbers for population, DC had a higher rate than any other state, at 96.69 offenses per million residents. Using the FBI rate method, this rate would be 9.67 reported offenses per 100,000 residents.  As the above graph shows, this hate crime rate corresponds with DC’s high rate of hate groups.  However, this relationship does not hold up when DC is excluded from the analysis, as can be seen in the graph below.  If DC is excluded, there is no statistically significant relationship between the concentration of hate groups and hate crimes in any of the other states, with only 2% of the variability accounted for.



Image: Paul Ricci.
 
Comparison of DC with New York City

So what factors besides poverty could be driving this relationship?  Compared to the other states DC has the highest population density by far at 11,157.58 persons per square mile.  Because Washington, DC is a quasi-city state, it may be appropriate to compare it to the US’s largest city, New York City (NYC).  In 2015, NYC had 8,550,405 inhabitants over a total of 302.64 square miles (approximately 488.13 km2) giving the city a population density of 28,252.72 persons per square mile.  I don’t have hate crime data for NYC but I can estimate the hate group rate from the hate group map of the Southern Poverty Law Center.  I counted 36 hate groups in the area, which would give NYC a rate of 4.21 groups per million – a number which is considerably below DC’s rate of 30.83 groups per million. In 2010, 25.5% of NYCs population identified as African-American whereas 50.7% of DCs population did.  Of the 21 total hate groups in DC, six of them are black separatist groups such as the Nation of Islam (28.6%).  Of the 36 hate groups in NYC, eight are black separatist (22.2%).  You can scan the other hate groups in each city here.

Looking at other statistics for NYC, the violent crime rate is 596.7 offenses per 100,000 residents and the murder rate sits at 3.9 offenses per 100,000 residents.  These are considerably lower than DCs rates of 1,244.4 violent offenses per 100,000 residents and 15.9 murders per 100,000 residents.  DC has a higher median household income at $70,848 than NYC’s $53,373.  Correspondingly, the 20.0% poverty rate for NYC is higher than DC’s 17.4%. 

Conclusion
 
One must be careful to draw grand conclusions from statistics that compare DC to the rest of the US and DC to NYC.  One can look at the obvious differences DC has with the other states. While it has three votes in the Electoral College for President, it has no members in Congress with full voting privileges on laws which may affect them. Further, as John Oliver explains, they have to pay full federal taxes:



We see Washington, DC portrayed in the media all the time but do we really notice what goes on there outside of the White House, the Capitol Building, and the various other federal buildings?  DC residents have been campaigning for full statehood for years but it has been stalled in Congress.  This second class citizenship may or may not explain all of the statistical discrepancies for DC.  The issue definitely merits further study.  There could be many other anomalies regarding DC of which I am not aware.

**Related Posts**

Don’t test me: Using Fisher’s exact test to unearth stories about statistical relationships (Repost) 

Concentration of Hate Groups Predict Hate Crimes (if you consider DC) and Trump Vote (if you don't)

SPLC Hate Group Update: Washington, DC has an Increase in Activity

Thursday, May 11, 2017

Paul Ricci, Independent for Johnstown, PA City Council

This is a cross post with my new website

Paul Ricci, Independent for Johnstown, PA City Council


Hello I'm Paul Ricci and I approve this message.

I was born here in 1970.  I have worked in a variety of fields and in a variety of cities which gives me a unique perspective on what works and what doesn’t for Johnstown.  Being a statistician - with experience in research methodology - helps me to see patterns in social trends that are not easy to see with the naked eye and to find solutions to problems when they are called for.  The platform I am running on is:
  • A Living Wage Ordinance for the City
  • End Corruption by Improving Enforcement & Transparency in City Government
  • City Wide Wifi
  • Consolidation of Local Communities with Johnstown
  • Fight the Drug Problem in the City through Better Treatment and Education
Our city has the third fastest shrinking population in the US, the graph below shows how our County, Cambria County, lags behind our state and the US as a whole in median household income, and our county ranked 63rd out of 67 PA counties in the annual County Health Rankings from the Robert Wood Johnson Foundation with low rankings for quality of life and length of life.


For our city to move forward, we need out-of-the-box thinkers as our leaders.  I am currently collecting signatures to be placed on the ballot for the November election.  I am running as an independent.  If anyone is interested in assisting my campaign or signing my petition (if you are a registered voter in Johnstown of any party affiliation or independent) you may contact me at riccipt@yahoo.com.

**Related Posts**

Trends in Cambria County Uninsured Show The Effect of Medicaid Expansion


2014 Income and Poverty Update for Pennsylvania and Cambria County


New Poverty Estimates for PA Counties

Tuesday, May 2, 2017

Education the Best Predictor for Trump Support in PA Counties

Trump marked the first 100 days milestone of his campaign with a rally in Harrisburg the State Capitol touting his achievements and his plan to repeal and replace the Affordable Care Act.  In my last post I looked at uninsured trends in the state.  This time I will look at uninsured, poverty rates, educational attainment rates and changes in substantiated child abuse rates and how they correlate with Trump's % of the vote in each county in 2016.


Map showing changes in substantiated child abuse cases from 2011 to 2015 in PA Counties.  From CNHI News and the PA DHS Child Abuse Report (The map was wrinkled in my folder)
The map above shows changes in substantiated Child abuse cases from 2011 to 2015 with the counties in red showing a 275% or more increase in cases.  Statewide there was an increase of 23.3% in cases.  This variable plus the other variables were considered in the correlation table below.  The numbers in the table below are correlation coefficients with significant ones in bold.  A positive coefficient of means that as one variable increases the other has a corresponding increase.  A negative correlation of means that for every increase in one variable there is a corresponding decrease in the other.  


Corr w Philly
Uninsured: %
% in poverty
% increase in Child Abuse
Bachelors
Some college
HS Grad
HS Only
LT High School
HS comp
Trump %
Uninsured: %
1









% in poverty
0.34
1








% increase in Child Abuse
-0.03
0.10
1







Bachelors
-0.41
-0.45
-0.16
1






Some college
-0.49
-0.47
-0.13
0.86
1





HS Grad
-0.73
-0.45
0.03
0.63
0.78
1




HS Only
0.35
0.42
0.18
-0.85
-0.97
-0.61
1



LT High School
0.73
0.45
-0.03
-0.63
-0.78
-1.00
0.61
1


HS comp
0.03
0.03
-0.04
-0.06
-0.13
0.00
0.17
0.00
1

Trump %
0.14
0.03
0.19
-0.77
-0.66
-0.31
0.71
0.31
0.28
1


The variables that are most strongly associated with Trump's % of the vote in each county are the educational attainment variables.  The strongest negative correlation with Trump's vote is the % of the county with a bachelor's degree or higher with a coefficient of -0.77.  The scatter plot shows that as the % with a bachelor's or higher increases, Trump's % of the vote decreases accounting for 58.4% of the variability in this relationship (if 100% of variability were accounted for, all of the counties would form a perfect straight line).  The one county that the model does not fit very well is Philadelphia which had 15% of the vote for Trump but had 25% with a bachelor's degree or higher which is close to the state rate.  To look at the effect that Philadelphia County has I recreated the table and scatterplot with it excluded.  27 pairwise correlations are significant above.



The correlation table below shows more significant relationships for Trump's % of the vote with the uninsured and poverty rates having a a weakly positive relationship of 0.33 and 0.30 respectively.  Looking at the % with a high school (HS) degree or higher there is a weak negative relationship (-0.52) but looking at the high school completion rate (those who enter HS in the 9th grade who finish) there is a positive relationship.  Doing some subtraction to find the % in each county who have a HS degree only and those with less than a HS degree the % in the county with a HS degree only has a strong positive relationship with a coefficient of 0.74 accounting for 55% of the variability.  Adding Philadelphia to the data had little effect on the correlation as can be seen in the scatterplot below.



Corr wo Philly
Uninsured: 
%
% in 
poverty
% increase in 
Child Abuse
Bachelors
Some 
college
HS 
Grad
HS 
Only
LT High 
School
HS 
comp
Trump %
Uninsured: %
1









% in poverty
0.24
1








% increase in Child Abuse
0.00
0.15
1







Bachelors
-0.44
-0.52
-0.15
1






Some coll
-0.52
-0.54
-0.13
0.86
1





HS Grad
-0.70
-0.39
0.01
0.67
0.82
1




HS Only
0.41
0.54
0.17
-0.86
-0.98
-0.68
1



LT High School
0.70
0.39
-0.01
-0.67
-0.82
-1.00
0.68
1


HS comp
0.04
0.04
-0.04
-0.05
-0.13
0.00
0.16
0.00
1

Trump %
0.33
0.30
0.17
-0.84
-0.73
-0.52
0.74
0.52
0.31
1


Philadelphia is an outlier for PA counties in much the same way as DC is for the rest of the states.  Child abuse case increases were not correlated with any other variables.  In the next post I will look at these variables in a multiple regression model.  Below is the raw data for the analysis. 

Name
Uninsured
%
% in
poverty
% increase in
Child Abuse
Bachelors
Or higher
Some
college
HS
Grad
HS
Only
LT High
School
HS
comp
Trump
%
Pennsylvania
7.6
13.1
23.30%
28.6
63
89
26.2
10.8
85
48.58
Adams County, PA
7.9
8.5
-40.3
21.7
53
87.3
34.8
12.7
92
66.17
Allegheny County, PA
6
12.2
14.8
37.8
77
93.5
16.3
6.5
89
39.91
Armstrong County, PA
7.2
12.6
16
15
54
88.9
34.6
11.1
89
74.27
Beaver County, PA
5.5
13.1
37.5
22.8
66
91.7
25.6
8.3
74
57.64
Bedford County, PA
8.5
13.9
275
13.5
47
86.0
38.7
14.0
92
82.72
Berks County, PA
8.6
12.6
17.7
23.2
57
85.1
28.4
14.9
84
52.78
Blair County, PA
6.4
15.3
12.5
19.6
54
90.5
36.2
9.5
88
71.27
Bradford County, PA
8.1
13
-4.3
17.6
46
88.5
42.7
11.5
89
70.57
Bucks County, PA
5.8
6.3
-31.7
37.4
72
93.5
21.7
6.5
92
47.74
Butler County, PA
4.9
9.5
-19.2
32.3
72
93.0
20.8
7.0
95
66.37
Cambria County, PA
6.6
14.9
65.2
19.1
58
89.6
31.6
10.4
90
67
Cameron County, PA
7.4
13.3
100
15.7
59
89.5
30.5
10.5
72.69
Carbon County, PA
7.1
11.5
9.5
15.5
52
88.7
36.9
11.3
81
65.13
Centre County, PA
7.9
16.1
110
41.4
72
93.2
21.3
6.8
92
46.32
Chester County, PA
6.4
6
-20.3
49.1
75
92.7
17.5
7.3
89
43.2
Clarion County, PA
8.1
17.4
322.2
20.1
56
88.7
33.1
11.3
94
71.72
Clearfield County, PA
7.3
16.6
41.9
13.2
45
87.1
41.7
12.9
89
72.75
Clinton County, PA
7.6
16
66.7
16.9
52
87.1
35.5
12.9
93
65.1
Columbia County, PA
6.7
15
18.8
21.1
54
88.5
34.0
11.5
88
63.78
Crawford County, PA
9.8
15.3
-21.2
19.8
52
87.7
35.4
12.3
90
66.65
Cumberland County, PA
6.8
7.3
102
32.7
66
91.5
26.0
8.5
90
56.8
Dauphin County, PA
7.4
13.6
146
28.4
63
88.8
25.8
11.2
84
46.51
Delaware County, PA
6.8
10.4
27
36
69
92.2
23.0
7.8
76
37.18
Elk County, PA
6
9.3
220
16.8
53
91.1
38.1
8.9
91
69.49
Erie County, PA
6.9
17.1
-18.3
26.1
59
90.8
31.3
9.2
87
48.57
Fayette County, PA
7.5
20.1
67.5
14.2
48
86.8
38.7
13.2
79
64.33
Forest County, PA
8.3
24.3
66.7
8.5
16
80.5
64.7
19.5
70.1
Franklin County, PA
9.4
9.4
23.4
19
50
86.0
35.8
14.0
86
71.37
Fulton County, PA
7.8
11.6
42.9
13.1
46
85.1
39.5
14.9
86
84.09
Greene County, PA
6.5
15.1
300
17.7
45
86.9
42.0
13.1
85
68.82
Huntingdon County, PA
7
14.3
80
14.3
44
88.8
45.1
11.2
89
73.55
Indiana County, PA
8.7
18.2
-4.3
22.6
58
88.5
30.4
11.5
94
65.89
Jefferson County, PA
8.8
14.6
58.3
14.6
49
88.6
39.3
11.4
92
78
Juniata County, PA
9.9
12.8
-81.3
13.1
37
81.7
44.8
18.3
94
79.14
Lackawanna County, PA
7.8
15.3
46.6
25.9
63
89.9
27.2
10.1
83
46.77
Lancaster County, PA
11.1
10.6
0
25.2
53
84.5
31.0
15.5
90
57.2
Lawrence County, PA
7.1
17.5
-16.7
19.7
60
89.2
29.5
10.8
93
62.4
Lebanon County, PA
8.7
11.6
89.7
19.6
52
85.7
33.5
14.3
86
65.53
Lehigh County, PA
8.6
12.1
18.8
28.5
62
87.4
25.0
12.6
84
45.77
Luzerne County, PA
7.8
15.1
52.1
21.4
59
88.9
29.8
11.1
87
58.29
Lycoming County, PA
7.2
14.8
46.6
20.4
57
88.0
31.3
12.0
87
70.46
Mc Kean County, PA
7.2
16.9
-20.7
16.7
48
89.7
41.7
10.3
90
71.4
Mercer County, PA
7.9
14.2
76.2
21.5
56
89.1
33.2
10.9
93
60.3
Mifflin County, PA
10.7
15.4
57.9
11.6
43
82.2
39.2
17.8
89
75.77
Monroe County, PA
8.5
12.7
12.7
23
64
89.7
26.2
10.3
89
47.86
Montgomery County, PA
5.3
6.6
8.6
46.9
78
93.8
15.8
6.2
94
37.44
Montour County, PA
5
9
50
28.8
61
89.4
28.3
10.6
88
61.8
Northampton County, PA
6.9
8.8
-10.7
27.2
66
89.8
23.6
10.2
73
49.98
Northumberland County, PA
7.3
13.1
23.8
14.6
45
85.6
40.6
14.4
87
69.43
Perry County, PA
9.3
9.6
0
16
55
88.5
33.4
11.5
91
73.81
Philadelphia County, PA
11
25.4
-9.2
25.4
58
82.0
23.7
18.0
88
15.37
Pike County, PA
8.6
10.9
200
23.7
63
91.0
28.3
9.0
70
61.51
Potter County, PA
7.8
14.3
0
15.7
51
87.8
37.2
12.2
92
80.31
Schuylkill County, PA
7.9
13
15.8
15.4
51
87.5
36.1
12.5
69.99
Snyder County, PA
10.5
11.7
38.9
16.5
45
82.4
37.7
17.6
89
71.66
Somerset County, PA
8.7
14.4
-36
15.3
47
86.5
39.8
13.5
87
76.54
Sullivan County, PA
8.9
13.5
300
16.1
47
89.6
42.7
10.4
93
73.05
Susquehanna County, PA
9.3
12.6
-17.2
17.1
49
89.1
40.5
10.9
68.34
Tioga County, PA
8.6
13.1
72.2
19.6
55
88.4
33.0
11.6
95
74.29
Union County, PA
8.9
12.4
45.5
20.5
46
83.9
37.5
16.1
88
60.81
Venango County, PA
7.3
13.5
22.2
15.8
50
89.0
39.0
11.0
95
68.62
Warren County, PA
7.1
12.2
-30.8
18.1
55
90.5
35.4
9.5
90
67.68
Washington County, PA
5.7
10.1
29.2
27.4
67
91.3
24.5
8.7
89
60.51
Wayne County, PA
8.1
13.3
-13.6
19.7
50
89.3
39.7
10.7
90
67.63
Westmoreland County, PA
5.9
11.3
-1.5
26.4
68
92.7
25.0
7.3
94
64.01
Wyoming County, PA
6.9
11.1
200
18.2
50
91.2
40.9
8.8
91
67.23
York County, PA
7.1
10.4
21.9
22.8
57
88.3
31.8
11.7
94
62.4
 

**Related Posts**

Trends in Cambria County Uninsured Show The Effect of Medicaid Expansion

 

2014 Income and Poverty Update for Pennsylvania and Cambria County

 

Hate Groups and Trump's Vote%: Predictive effect present when education and poverty are considered

 

More Hate Groups in States Where Trump and Clinton Win (and in DC Where He Lost)