Skip to content

Instantly share code, notes, and snippets.

@wmay

wmay/trump_hate_crimes.R

Last active September 10, 2019 20:49
Show Gist options
  • Save wmay/0c032d4bc7aca899388702a164c00326 to your computer and use it in GitHub Desktop.
Save wmay/0c032d4bc7aca899388702a164c00326 to your computer and use it in GitHub Desktop.
Download ZIP
Trump rallies and hate crimes
Raw
trump_hate_crimes.R
## are Trump (or Clinton) rallies associated with hate crimes?
library(MASS)
## organize data (downloaded from the analysis page, under the "Hate
## Incidents" tab)
## https://public.tableau.com/profile/matthew.lilley#!/vizhome/TrumpRallies_15676277431850/Population-TrumpRally
hate = read.csv('Hate_Incidents_data.csv', check.names = F,
stringsAsFactors = F)
hate$Geoid = as.factor(hate$Geoid)
hate$month = as.Date(hate$`Month-Year`, format = '%m/%d/%Y')
hate$log_pop = pmax(hate$`Trump Rally Population (Log)`,
hate$`No Trump Rally Population (Log)`)
## quick check, make sure I see a spike in November like described in
## the paper
monthly_hate = aggregate(`Hate Incidents` ~ month, FUN = sum,
data = hate)
plot(monthly_hate$month, monthly_hate$`Hate Incidents`, type = 'l',
xlab = 'Month', ylab = 'Hate Incidents')
## try to reproduce paper results
f_paper = `Hate Incidents` ~ `Trump Rally Occurred` +
`Jewish Population Share` + `Hate Groups` + `Violent Crime Rate` +
`Property Crime Rate` + `Romney 2012 Vote Share` +
`College Educated Share` + Region + as.factor(month)
r_paper = glm.nb(f_paper, hate, control = glm.control(maxit = 50))
summary(r_paper)
## pretty close!
## try to reproduce reason/Clinton results
f_reason1 = `Hate Incidents` ~ `Clinton Rally Occurred` +
`Jewish Population Share` + `Hate Groups` + `Violent Crime Rate` +
`Property Crime Rate` + `Romney 2012 Vote Share` +
`College Educated Share` + Region + as.factor(month)
r_reason1 = glm.nb(f_reason1, hate, control = glm.control(maxit = 50))
summary(r_reason1)
## pretty close!
## try to reproduce reason/population results
f_reason2 = `Hate Incidents` ~ `Trump Rally Occurred` +
`Clinton Rally Occurred` + log_pop + `Jewish Population Share` +
`Hate Groups` + `Violent Crime Rate` + `Property Crime Rate` +
`Romney 2012 Vote Share` + `College Educated Share` + Region +
as.factor(month)
r_reason2 = glm.nb(f_reason2, hate, control = glm.control(maxit = 50))
summary(r_reason2)
## pretty close!
## now try with county fixed effects substituting for county-level
## predictors
f_fe = `Hate Incidents` ~ `Trump Rally Occurred` +
`Clinton Rally Occurred` + as.factor(month) + Geoid
## r_fe = glm.nb(f_fe, hate, control = glm.control(maxit = 50))
## ^ can't run for this model, glm.nb is too slow
## try again, removing counties without rallies
rally_counties = unique(hate$Geoid[hate$`Trump Rally Occurred` |
hate$`Clinton Rally Occurred`])
hate2 = subset(hate, Geoid %in% rally_counties)
r_fe = glm.nb(f_fe, hate2, control = glm.control(maxit = 50))
summary(r_fe)
## finds smaller Trump rally association but greater
## uncertainty. Neither Trump nor Clinton rally effects are
## statistically significant
Raw
trump_hate_crimes_output.txt
R version 3.6.1 (2019-07-05) -- "Action of the Toes"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
Natural language support but running in an English locale
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> ## are Trump (or Clinton) rallies associated with hate crimes?
>
> library(MASS)
>
> ## organize data (downloaded from the analysis page, under the "Hate
> ## Incidents" tab)
> ## https://public.tableau.com/profile/matthew.lilley#!/vizhome/TrumpRallies_15676277431850/Population-TrumpRally
> hate = read.csv('Hate_Incidents_data.csv', check.names = F,
+ stringsAsFactors = F)
> hate$Geoid = as.factor(hate$Geoid)
> hate$month = as.Date(hate$`Month-Year`, format = '%m/%d/%Y')
> hate$log_pop = pmax(hate$`Trump Rally Population (Log)`,
+ hate$`No Trump Rally Population (Log)`)
>
> ## quick check, make sure I see a spike in November like described in
> ## the paper
> monthly_hate = aggregate(`Hate Incidents` ~ month, FUN = sum,
+ data = hate)
> plot(monthly_hate$month, monthly_hate$`Hate Incidents`, type = 'l',
+ xlab = 'Month', ylab = 'Hate Incidents')
>
> ## try to reproduce paper results
> f_paper = `Hate Incidents` ~ `Trump Rally Occurred` +
+ `Jewish Population Share` + `Hate Groups` + `Violent Crime Rate` +
+ `Property Crime Rate` + `Romney 2012 Vote Share` +
+ `College Educated Share` + Region + as.factor(month)
> r_paper = glm.nb(f_paper, hate, control = glm.control(maxit = 50))
> summary(r_paper)
Call:
glm.nb(formula = f_paper, data = hate, control = glm.control(maxit = 50),
init.theta = 0.2345175165, link = log)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.9695 -0.1229 -0.0794 -0.0588 5.5398
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -5.2421736 0.3230929 -16.225 < 2e-16 ***
`Trump Rally Occurred` 1.1126400 0.1156827 9.618 < 2e-16 ***
`Jewish Population Share` 30.2905198 1.3304257 22.768 < 2e-16 ***
`Hate Groups` 0.0219351 0.0021908 10.012 < 2e-16 ***
`Violent Crime Rate` 0.0042928 0.0010688 4.017 5.91e-05 ***
`Property Crime Rate` 0.0007101 0.0002907 2.443 0.01456 *
`Romney 2012 Vote Share` -4.0780470 0.3494679 -11.669 < 2e-16 ***
`College Educated Share` 5.5039034 0.4161534 13.226 < 2e-16 ***
RegionNortheast 1.0950162 0.1436050 7.625 2.44e-14 ***
RegionSouth -0.2919943 0.1488601 -1.962 0.04982 *
RegionWest 0.4188024 0.1599306 2.619 0.00883 **
as.factor(month)2016-02-01 -0.0309747 0.2536446 -0.122 0.90281
as.factor(month)2016-03-01 0.5793208 0.2311355 2.506 0.01220 *
as.factor(month)2016-04-01 0.2301420 0.2411973 0.954 0.34000
as.factor(month)2016-05-01 0.2536822 0.2400133 1.057 0.29053
as.factor(month)2016-06-01 0.2168765 0.2410137 0.900 0.36820
as.factor(month)2016-07-01 -0.1999036 0.2564044 -0.780 0.43560
as.factor(month)2016-08-01 0.0234396 0.2469937 0.095 0.92439
as.factor(month)2016-09-01 0.0484270 0.2457390 0.197 0.84378
as.factor(month)2016-10-01 0.3252195 0.2361296 1.377 0.16842
as.factor(month)2016-11-01 0.9779578 0.2196186 4.453 8.47e-06 ***
as.factor(month)2016-12-01 0.5416994 0.2299069 2.356 0.01846 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for Negative Binomial(0.2345) family taken to be 1)
Null deviance: 6891.1 on 37667 degrees of freedom
Residual deviance: 2521.8 on 37646 degrees of freedom
(36 observations deleted due to missingness)
AIC: 5868.3
Number of Fisher Scoring iterations: 1
Theta: 0.2345
Std. Err.: 0.0163
2 x log-likelihood: -5822.3130
> ## pretty close!
>
> ## try to reproduce reason/Clinton results
> f_reason1 = `Hate Incidents` ~ `Clinton Rally Occurred` +
+ `Jewish Population Share` + `Hate Groups` + `Violent Crime Rate` +
+ `Property Crime Rate` + `Romney 2012 Vote Share` +
+ `College Educated Share` + Region + as.factor(month)
> r_reason1 = glm.nb(f_reason1, hate, control = glm.control(maxit = 50))
> summary(r_reason1)
Call:
glm.nb(formula = f_reason1, data = hate, control = glm.control(maxit = 50),
init.theta = 0.2127625009, link = log)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.9169 -0.1253 -0.0811 -0.0595 5.2423
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -5.6560349 0.3283270 -17.227 < 2e-16 ***
`Clinton Rally Occurred` 1.2019769 0.2479582 4.847 1.25e-06 ***
`Jewish Population Share` 30.2364876 1.3723678 22.032 < 2e-16 ***
`Hate Groups` 0.0220490 0.0022284 9.895 < 2e-16 ***
`Violent Crime Rate` 0.0049403 0.0009568 5.163 2.43e-07 ***
`Property Crime Rate` 0.0009619 0.0002473 3.890 0.00010 ***
`Romney 2012 Vote Share` -3.8853314 0.3499793 -11.102 < 2e-16 ***
`College Educated Share` 6.5095567 0.4140531 15.722 < 2e-16 ***
RegionNortheast 1.0783948 0.1448708 7.444 9.78e-14 ***
RegionSouth -0.3596049 0.1491072 -2.412 0.01588 *
RegionWest 0.3388020 0.1585812 2.136 0.03264 *
as.factor(month)2016-02-01 -0.0002848 0.2585157 -0.001 0.99912
as.factor(month)2016-03-01 0.6465374 0.2358017 2.742 0.00611 **
as.factor(month)2016-04-01 0.3598494 0.2448748 1.470 0.14169
as.factor(month)2016-05-01 0.3858107 0.2439913 1.581 0.11382
as.factor(month)2016-06-01 0.3817982 0.2441271 1.564 0.11783
as.factor(month)2016-07-01 -0.0290382 0.2591198 -0.112 0.91077
as.factor(month)2016-08-01 0.1650499 0.2510312 0.657 0.51087
as.factor(month)2016-09-01 0.2340220 0.2482450 0.943 0.34583
as.factor(month)2016-10-01 0.5177183 0.2388808 2.167 0.03021 *
as.factor(month)2016-11-01 1.1802676 0.2223231 5.309 1.10e-07 ***
as.factor(month)2016-12-01 0.7562779 0.2321419 3.258 0.00112 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for Negative Binomial(0.2128) family taken to be 1)
Null deviance: 6713.9 on 37667 degrees of freedom
Residual deviance: 2498.3 on 37646 degrees of freedom
(36 observations deleted due to missingness)
AIC: 5930.7
Number of Fisher Scoring iterations: 1
Theta: 0.2128
Std. Err.: 0.0144
2 x log-likelihood: -5884.7350
> ## pretty close!
>
> ## try to reproduce reason/population results
> f_reason2 = `Hate Incidents` ~ `Trump Rally Occurred` +
+ `Clinton Rally Occurred` + log_pop + `Jewish Population Share` +
+ `Hate Groups` + `Violent Crime Rate` + `Property Crime Rate` +
+ `Romney 2012 Vote Share` + `College Educated Share` + Region +
+ as.factor(month)
> r_reason2 = glm.nb(f_reason2, hate, control = glm.control(maxit = 50))
> summary(r_reason2)
Call:
glm.nb(formula = f_reason2, data = hate, control = glm.control(maxit = 50),
init.theta = 0.7382578965, link = log)
Deviance Residuals:
Min 1Q Median 3Q Max
-2.2491 -0.0921 -0.0447 -0.0255 5.9145
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.903e+01 6.571e-01 -28.959 < 2e-16 ***
`Trump Rally Occurred` 1.675e-01 1.000e-01 1.676 0.093825 .
`Clinton Rally Occurred` 1.018e-01 1.878e-01 0.542 0.587970
log_pop 1.173e+00 4.406e-02 26.614 < 2e-16 ***
`Jewish Population Share` 9.752e+00 1.022e+00 9.545 < 2e-16 ***
`Hate Groups` -5.409e-03 2.050e-03 -2.638 0.008327 **
`Violent Crime Rate` 2.518e-03 1.997e-03 1.261 0.207283
`Property Crime Rate` 8.116e-04 5.590e-04 1.452 0.146526
`Romney 2012 Vote Share` -1.334e+00 3.515e-01 -3.796 0.000147 ***
`College Educated Share` 3.035e+00 4.385e-01 6.920 4.53e-12 ***
RegionNortheast 1.136e+00 1.342e-01 8.463 < 2e-16 ***
RegionSouth 1.620e-01 1.374e-01 1.180 0.238191
RegionWest 3.924e-01 1.488e-01 2.638 0.008339 **
as.factor(month)2016-02-01 -5.752e-02 2.217e-01 -0.259 0.795312
as.factor(month)2016-03-01 5.883e-01 2.008e-01 2.930 0.003390 **
as.factor(month)2016-04-01 2.502e-01 2.107e-01 1.187 0.235092
as.factor(month)2016-05-01 2.823e-01 2.098e-01 1.346 0.178454
as.factor(month)2016-06-01 1.668e-01 2.136e-01 0.781 0.435032
as.factor(month)2016-07-01 -1.677e-01 2.260e-01 -0.742 0.457936
as.factor(month)2016-08-01 6.837e-02 2.171e-01 0.315 0.752838
as.factor(month)2016-09-01 1.582e-01 2.143e-01 0.738 0.460432
as.factor(month)2016-10-01 4.877e-01 2.050e-01 2.379 0.017363 *
as.factor(month)2016-11-01 1.129e+00 1.915e-01 5.893 3.80e-09 ***
as.factor(month)2016-12-01 6.713e-01 2.007e-01 3.345 0.000824 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for Negative Binomial(0.7383) family taken to be 1)
Null deviance: 8772.4 on 37667 degrees of freedom
Residual deviance: 2539.6 on 37644 degrees of freedom
(36 observations deleted due to missingness)
AIC: 5068.2
Number of Fisher Scoring iterations: 1
Theta: 0.7383
Std. Err.: 0.0754
2 x log-likelihood: -5018.1970
> ## pretty close!
>
>
> ## now try with county fixed effects substituting for county-level
> ## predictors
>
> f_fe = `Hate Incidents` ~ `Trump Rally Occurred` +
+ `Clinton Rally Occurred` + as.factor(month) + Geoid
> ## r_fe = glm.nb(f_fe, hate, control = glm.control(maxit = 50))
> ## ^ can't run for this model, glm.nb is too slow
>
> ## try again, removing counties without rallies
> rally_counties = unique(hate$Geoid[hate$`Trump Rally Occurred` |
+ hate$`Clinton Rally Occurred`])
> hate2 = subset(hate, Geoid %in% rally_counties)
> r_fe = glm.nb(f_fe, hate2, control = glm.control(maxit = 50))
> summary(r_fe)
Call:
glm.nb(formula = f_fe, data = hate2, control = glm.control(maxit = 50),
init.theta = 43.45759162, link = log)
Deviance Residuals:
Min 1Q Median 3Q Max
-2.3639 -0.3414 0.0000 0.0000 2.7873
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -5.246e+01 1.937e+07 0.000 1.00000
`Trump Rally Occurred` 6.863e-02 1.641e-01 0.418 0.67584
`Clinton Rally Occurred` -1.462e-01 1.833e-01 -0.798 0.42496
as.factor(month)2016-02-01 -3.388e-02 2.646e-01 -0.128 0.89811
as.factor(month)2016-03-01 5.773e-01 2.349e-01 2.458 0.01399 *
as.factor(month)2016-04-01 4.285e-01 2.474e-01 1.732 0.08324 .
as.factor(month)2016-05-01 1.244e-01 2.662e-01 0.467 0.64017
as.factor(month)2016-06-01 -9.716e-02 2.800e-01 -0.347 0.72860
as.factor(month)2016-07-01 -2.063e-01 2.904e-01 -0.710 0.47754
as.factor(month)2016-08-01 -2.389e-01 2.979e-01 -0.802 0.42253
as.factor(month)2016-09-01 8.578e-02 2.852e-01 0.301 0.76362
as.factor(month)2016-10-01 5.448e-01 2.744e-01 1.986 0.04708 *
as.factor(month)2016-11-01 1.319e+00 2.574e-01 5.125 2.98e-07 ***
as.factor(month)2016-12-01 8.574e-01 2.676e-01 3.205 0.00135 **
Geoid1097 5.026e+01 1.937e+07 0.000 1.00000
Geoid4013 5.151e+01 1.937e+07 0.000 1.00000
Geoid4019 5.062e+01 1.937e+07 0.000 1.00000
Geoid4025 1.512e+01 2.734e+07 0.000 1.00000
Geoid5007 1.507e+01 2.737e+07 0.000 1.00000
Geoid5119 1.507e+01 2.737e+07 0.000 1.00000
Geoid6019 1.509e+01 2.736e+07 0.000 1.00000
Geoid6059 5.114e+01 1.937e+07 0.000 1.00000
Geoid6067 1.509e+01 2.736e+07 0.000 1.00000
Geoid6073 5.297e+01 1.937e+07 0.000 1.00000
Geoid6085 5.161e+01 1.937e+07 0.000 1.00000
Geoid6089 1.509e+01 2.736e+07 0.000 1.00000
Geoid8001 1.519e+01 2.740e+07 0.000 1.00000
Geoid8031 5.267e+01 1.937e+07 0.000 1.00000
Geoid8041 4.954e+01 1.937e+07 0.000 1.00000
Geoid8059 5.064e+01 1.937e+07 0.000 1.00000
Geoid8069 5.024e+01 1.937e+07 0.000 1.00000
Geoid8077 1.512e+01 2.734e+07 0.000 1.00000
Geoid8101 1.515e+01 2.740e+07 0.000 1.00000
Geoid8123 5.093e+01 1.937e+07 0.000 1.00000
Geoid9001 5.209e+01 1.937e+07 0.000 1.00000
Geoid9003 5.131e+01 1.937e+07 0.000 1.00000
Geoid9009 5.216e+01 1.937e+07 0.000 1.00000
Geoid10001 1.508e+01 2.736e+07 0.000 1.00000
Geoid12005 1.512e+01 2.734e+07 0.000 1.00000
Geoid12009 5.023e+01 1.937e+07 0.000 1.00000
Geoid12011 5.295e+01 1.937e+07 0.000 1.00000
Geoid12021 5.115e+01 1.937e+07 0.000 1.00000
Geoid12031 5.023e+01 1.937e+07 0.000 1.00000
Geoid12033 4.951e+01 1.937e+07 0.000 1.00000
Geoid12057 5.068e+01 1.937e+07 0.000 1.00000
Geoid12071 5.024e+01 1.937e+07 0.000 1.00000
Geoid12073 5.115e+01 1.937e+07 0.000 1.00000
Geoid12083 1.512e+01 2.734e+07 0.000 1.00000
Geoid12086 5.282e+01 1.937e+07 0.000 1.00000
Geoid12095 5.097e+01 1.937e+07 0.000 1.00000
Geoid12097 5.031e+01 1.937e+07 0.000 1.00000
Geoid12099 5.332e+01 1.937e+07 0.000 1.00000
Geoid12101 1.516e+01 2.740e+07 0.000 1.00000
Geoid12103 5.103e+01 1.937e+07 0.000 1.00000
Geoid12105 4.955e+01 1.937e+07 0.000 1.00000
Geoid12109 4.955e+01 1.937e+07 0.000 1.00000
Geoid12111 1.518e+01 2.740e+07 0.000 1.00000
Geoid12115 4.955e+01 1.937e+07 0.000 1.00000
Geoid12117 4.959e+01 1.937e+07 0.000 1.00000
Geoid12127 5.030e+01 1.937e+07 0.000 1.00000
Geoid13121 5.281e+01 1.937e+07 0.000 1.00000
Geoid13185 1.507e+01 2.737e+07 0.000 1.00000
Geoid17113 1.508e+01 2.737e+07 0.000 1.00000
Geoid18003 5.022e+01 1.937e+07 0.000 1.00000
Geoid18057 1.509e+01 2.736e+07 0.000 1.00000
Geoid18097 5.021e+01 1.937e+07 0.000 1.00000
Geoid18141 1.509e+01 2.736e+07 0.000 1.00000
Geoid18163 1.508e+01 2.736e+07 0.000 1.00000
Geoid18167 1.509e+01 2.736e+07 0.000 1.00000
Geoid19013 1.507e+01 2.737e+07 0.000 1.00000
Geoid19033 1.507e+01 2.737e+07 0.000 1.00000
Geoid19045 1.507e+01 2.737e+07 0.000 1.00000
Geoid19061 1.507e+01 2.737e+07 0.000 1.00000
Geoid19103 1.507e+01 2.737e+07 0.000 1.00000
Geoid19113 1.511e+01 2.740e+07 0.000 1.00000
Geoid19125 1.507e+01 2.737e+07 0.000 1.00000
Geoid19127 1.507e+01 2.737e+07 0.000 1.00000
Geoid19139 1.507e+01 2.737e+07 0.000 1.00000
Geoid19153 1.513e+01 2.740e+07 0.000 1.00000
Geoid19155 1.507e+01 2.737e+07 0.000 1.00000
Geoid19163 1.507e+01 2.737e+07 0.000 1.00000
Geoid19167 1.507e+01 2.737e+07 0.000 1.00000
Geoid19169 1.507e+01 2.737e+07 0.000 1.00000
Geoid19179 1.507e+01 2.737e+07 0.000 1.00000
Geoid19181 1.507e+01 2.737e+07 0.000 1.00000
Geoid19193 1.512e+01 2.734e+07 0.000 1.00000
Geoid20173 1.508e+01 2.737e+07 0.000 1.00000
Geoid21111 1.508e+01 2.737e+07 0.000 1.00000
Geoid22033 4.951e+01 1.937e+07 0.000 1.00000
Geoid22071 4.951e+01 1.937e+07 0.000 1.00000
Geoid23001 1.512e+01 2.734e+07 0.000 1.00000
Geoid23005 5.090e+01 1.937e+07 0.000 1.00000
Geoid23019 1.509e+01 2.736e+07 0.000 1.00000
Geoid24043 1.508e+01 2.736e+07 0.000 1.00000
Geoid24047 1.508e+01 2.736e+07 0.000 1.00000
Geoid25017 5.345e+01 1.937e+07 0.000 1.00000
Geoid26045 1.511e+01 2.735e+07 0.000 1.00000
Geoid26081 1.512e+01 2.734e+07 0.000 1.00000
Geoid26099 1.508e+01 2.737e+07 0.000 1.00000
Geoid26125 5.161e+01 1.937e+07 0.000 1.00000
Geoid26139 4.962e+01 1.937e+07 0.000 1.00000
Geoid26163 5.075e+01 1.937e+07 0.000 1.00000
Geoid26165 1.508e+01 2.737e+07 0.000 1.00000
Geoid27053 5.174e+01 1.937e+07 0.000 1.00000
Geoid28047 1.507e+01 2.737e+07 0.000 1.00000
Geoid28049 1.511e+01 2.735e+07 0.000 1.00000
Geoid28089 1.508e+01 2.737e+07 0.000 1.00000
Geoid29095 1.508e+01 2.737e+07 0.000 1.00000
Geoid29510 5.091e+01 1.937e+07 0.000 1.00000
Geoid30111 1.509e+01 2.736e+07 0.000 1.00000
Geoid31055 5.191e+01 1.937e+07 0.000 1.00000
Geoid32003 5.028e+01 1.937e+07 0.000 1.00000
Geoid32031 1.513e+01 2.740e+07 0.000 1.00000
Geoid33001 1.511e+01 2.735e+07 0.000 1.00000
Geoid33009 1.507e+01 2.737e+07 0.000 1.00000
Geoid33011 5.067e+01 1.937e+07 0.000 1.00000
Geoid33013 4.951e+01 1.937e+07 0.000 1.00000
Geoid33015 5.020e+01 1.937e+07 0.000 1.00000
Geoid33017 4.958e+01 1.937e+07 0.000 1.00000
Geoid33019 4.951e+01 1.937e+07 0.000 1.00000
Geoid35001 5.147e+01 1.937e+07 0.000 1.00000
Geoid36001 4.952e+01 1.937e+07 0.000 1.00000
Geoid36019 1.508e+01 2.736e+07 0.000 1.00000
Geoid36027 5.021e+01 1.937e+07 0.000 1.00000
Geoid36029 5.022e+01 1.937e+07 0.000 1.00000
Geoid36045 1.508e+01 2.736e+07 0.000 1.00000
Geoid36055 4.952e+01 1.937e+07 0.000 1.00000
Geoid36059 5.323e+01 1.937e+07 0.000 1.00000
Geoid36065 1.508e+01 2.736e+07 0.000 1.00000
Geoid36067 1.508e+01 2.736e+07 0.000 1.00000
Geoid37021 4.954e+01 1.937e+07 0.000 1.00000
Geoid37025 1.508e+01 2.737e+07 0.000 1.00000
Geoid37051 1.508e+01 2.737e+07 0.000 1.00000
Geoid37061 1.511e+01 2.735e+07 0.000 1.00000
Geoid37067 1.514e+01 2.740e+07 0.000 1.00000
Geoid37081 1.514e+01 2.740e+07 0.000 1.00000
Geoid37089 1.512e+01 2.734e+07 0.000 1.00000
Geoid37101 1.512e+01 2.734e+07 0.000 1.00000
Geoid37107 1.512e+01 2.734e+07 0.000 1.00000
Geoid37119 5.031e+01 1.937e+07 0.000 1.00000
Geoid37129 1.511e+01 2.735e+07 0.000 1.00000
Geoid37147 1.514e+01 2.740e+07 0.000 1.00000
Geoid37183 1.515e+01 2.740e+07 0.000 1.00000
Geoid39007 1.512e+01 2.734e+07 0.000 1.00000
Geoid39013 1.509e+01 2.736e+07 0.000 1.00000
Geoid39023 1.512e+01 2.734e+07 0.000 1.00000
Geoid39027 1.511e+01 2.735e+07 0.000 1.00000
Geoid39035 5.154e+01 1.937e+07 0.000 1.00000
Geoid39041 1.512e+01 2.734e+07 0.000 1.00000
Geoid39049 5.099e+01 1.937e+07 0.000 1.00000
Geoid39061 1.514e+01 2.740e+07 0.000 1.00000
Geoid39095 1.514e+01 2.740e+07 0.000 1.00000
Geoid39099 1.521e+01 2.740e+07 0.000 1.00000
Geoid39113 5.021e+01 1.937e+07 0.000 1.00000
Geoid39133 1.517e+01 2.740e+07 0.000 1.00000
Geoid39151 4.954e+01 1.937e+07 0.000 1.00000
Geoid39153 1.511e+01 2.735e+07 0.000 1.00000
Geoid39155 1.508e+01 2.737e+07 0.000 1.00000
Geoid40109 1.507e+01 2.737e+07 0.000 1.00000
Geoid40143 4.951e+01 1.937e+07 0.000 1.00000
Geoid41039 1.509e+01 2.736e+07 0.000 1.00000
Geoid42003 4.961e+01 1.937e+07 0.000 1.00000
Geoid42007 1.512e+01 2.734e+07 0.000 1.00000
Geoid42013 1.511e+01 2.735e+07 0.000 1.00000
Geoid42017 5.115e+01 1.937e+07 0.000 1.00000
Geoid42021 1.519e+01 2.739e+07 0.000 1.00000
Geoid42029 1.508e+01 2.736e+07 0.000 1.00000
Geoid42041 1.511e+01 2.735e+07 0.000 1.00000
Geoid42043 5.030e+01 1.937e+07 0.000 1.00000
Geoid42045 5.149e+01 1.937e+07 0.000 1.00000
Geoid42049 1.511e+01 2.735e+07 0.000 1.00000
Geoid42069 1.516e+01 2.740e+07 0.000 1.00000
Geoid42071 5.064e+01 1.937e+07 0.000 1.00000
Geoid42079 1.508e+01 2.736e+07 0.000 1.00000
Geoid42091 5.161e+01 1.937e+07 0.000 1.00000
Geoid42101 5.260e+01 1.937e+07 0.000 1.00000
Geoid44003 1.508e+01 2.736e+07 0.000 1.00000
Geoid45003 1.507e+01 2.737e+07 0.000 1.00000
Geoid45007 1.507e+01 2.737e+07 0.000 1.00000
Geoid45013 1.507e+01 2.737e+07 0.000 1.00000
Geoid45019 5.090e+01 1.937e+07 0.000 1.00000
Geoid45021 1.507e+01 2.737e+07 0.000 1.00000
Geoid45029 1.507e+01 2.737e+07 0.000 1.00000
Geoid45041 1.507e+01 2.737e+07 0.000 1.00000
Geoid45043 1.507e+01 2.737e+07 0.000 1.00000
Geoid45045 4.951e+01 1.937e+07 0.000 1.00000
Geoid45051 4.951e+01 1.937e+07 0.000 1.00000
Geoid45063 1.507e+01 2.737e+07 0.000 1.00000
Geoid45085 1.507e+01 2.737e+07 0.000 1.00000
Geoid45091 1.507e+01 2.737e+07 0.000 1.00000
Geoid47157 1.507e+01 2.737e+07 0.000 1.00000
Geoid48113 4.953e+01 1.937e+07 0.000 1.00000
Geoid48339 1.509e+01 2.736e+07 0.000 1.00000
Geoid48439 1.507e+01 2.737e+07 0.000 1.00000
Geoid48453 5.148e+01 1.937e+07 0.000 1.00000
Geoid49035 5.021e+01 1.937e+07 0.000 1.00000
Geoid50007 5.020e+01 1.937e+07 0.000 1.00000
Geoid51107 4.954e+01 1.937e+07 0.000 1.00000
Geoid51630 1.511e+01 2.735e+07 0.000 1.00000
Geoid51680 1.507e+01 2.737e+07 0.000 1.00000
Geoid51750 1.507e+01 2.737e+07 0.000 1.00000
Geoid51760 4.953e+01 1.937e+07 0.000 1.00000
Geoid51770 1.511e+01 2.735e+07 0.000 1.00000
Geoid51810 1.512e+01 2.734e+07 0.000 1.00000
Geoid53061 1.511e+01 2.735e+07 0.000 1.00000
Geoid53063 1.509e+01 2.736e+07 0.000 1.00000
Geoid53073 1.509e+01 2.736e+07 0.000 1.00000
Geoid54039 1.509e+01 2.736e+07 0.000 1.00000
Geoid55009 1.508e+01 2.737e+07 0.000 1.00000
Geoid55031 1.508e+01 2.736e+07 0.000 1.00000
Geoid55035 1.508e+01 2.736e+07 0.000 1.00000
Geoid55063 1.508e+01 2.736e+07 0.000 1.00000
Geoid55073 1.508e+01 2.736e+07 0.000 1.00000
Geoid55079 5.241e+01 1.937e+07 0.000 1.00000
Geoid55087 1.508e+01 2.737e+07 0.000 1.00000
Geoid55101 1.508e+01 2.736e+07 0.000 1.00000
Geoid55105 1.508e+01 2.737e+07 0.000 1.00000
Geoid55131 1.511e+01 2.735e+07 0.000 1.00000
Geoid55133 1.511e+01 2.735e+07 0.000 1.00000
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for Negative Binomial(43.4576) family taken to be 1)
Null deviance: 2594.56 on 2471 degrees of freedom
Residual deviance: 699.32 on 2253 degrees of freedom
AIC: 1821.5
Number of Fisher Scoring iterations: 1
Theta: 43.5
Std. Err.: 77.1
2 x log-likelihood: -1381.475
> ## finds smaller Trump rally association but greater
> ## uncertainty. Neither Trump nor Clinton rally effects are
> ## statistically significant
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment