Problem Set 1

getwd()  #see where R is pulling data from

## [1] "C:/Users/soumi/OneDrive/Documents"

setwd("~/Desktop/PLSC 309/ProblemSets")  #tell R where to read data from
mydata <- read.csv("StateTurnout2020.csv") #the data set

Part 1: Learn about the dataset

Question 1

nrow(mydata) #Number of observations (rows)

## [1] 51

ncol(mydata) #Number of Variables

## [1] 35

head(mydata) #First 6 rows of data

##        State Ballots_cast    VEP_pop  VAP_pop Registered_total
## 1    ALABAMA      2323282  3761001.3  3917288             2527
## 2     ALASKA       359530   528179.7   553820              383
## 3    ARIZONA      3387054  4984557.4  5574070             3878
## 4   ARKANSAS      1219069  2166786.5  2315001             1361
## 5 CALIFORNIA     17500881 25917882.1 30730598            18001
## 6   COLORADO      3256980  4246894.6  4525319             2993
##   Reported_votes_total VEP_pop_18_24 VEP_pop_25_34 VEP_pop_35_44 VEP_pop_45_64
## 1                 2247           427           587           608          1246
## 2                  330            59            94            99           174
## 3                 3649           629           850           756          1627
## 4                 1186           261           389           319           680
## 5                16893          3334          4957          4016          7973
## 6                 2837           481           942           629          1302
##   VEP_pop_65_plus VEP_pop_Female VEP_pop_Latino VEP_pop_Male VEP_pop_Total
## 1             848           1960             53         1755          3716
## 2              91            253             27          264           516
## 3            1213           2610           1340         2465          5075
## 4             547           1138             83         1057          2195
## 5            5666          13366           8305        12580         25946
## 6             845           2124            618         2076          4200
##   VEP_pop_White Reported_votes_18_24 Reported_votes_25_34 Reported_votes_35_44
## 1          2569                  194                  315                  365
## 2           323                   29                   41                   62
## 3          3096                  327                  579                  509
## 4          1733                   84                  176                  179
## 5         11685                 1786                 2976                 2554
## 6          3220                  257                  608                  420
##   Reported_votes_45_64 Reported_votes_65_plus Reported_votes_female
## 1                  824                    549                  1209
## 2                  131                     67                   165
## 3                 1242                    992                  1996
## 4                  401                    346                   640
## 5                 5534                   4044                  8882
## 6                  922                    631                  1482
##   Reported_votes_latino Reported_votes_male Reported_votes_white VAP_pop_18_24
## 1                    30                1038                 1617           431
## 2                    17                 165                  230            60
## 3                   814                1653                 2385           653
## 4                    29                 546                  988           263
## 5                  4539                8012                 8711          3685
## 6                   315                1355                 2316           528
##   VAP_pop_25_34 VAP_pop_35_44 VAP_pop_45_64 VAP_pop_65_plus VAP_pop_Female
## 1           599           626          1261             852           1990
## 2            97           102           177              92            259
## 3           941           929          1833            1282           2899
## 4           420           341           706             553           1182
## 5          5953          5090          9499            6117          15556
## 6           996           717          1432             852           2271
##   VAP_pop_Latino VAP_pop_Male VAP_pop_census VAP_pop_White
## 1             79         1780           3769          2587
## 2             28          269            528           325
## 3           1800         2739           5638          3140
## 4            134         1101           2283          1744
## 5          11165        14786          30342         12090
## 6            854         2254           4525          3267

The first number, 51 corresponds to the number of observations or rows in the data set. The second number, 35 corresponds to the number of variables or columns in the data set.

Question 2

colSums(is.na(mydata)) #Number of missing values for each column

##                  State           Ballots_cast                VEP_pop 
##                      0                      0                      0 
##                VAP_pop       Registered_total   Reported_votes_total 
##                      0                      0                      0 
##          VEP_pop_18_24          VEP_pop_25_34          VEP_pop_35_44 
##                      0                      0                      0 
##          VEP_pop_45_64        VEP_pop_65_plus         VEP_pop_Female 
##                      0                      0                      0 
##         VEP_pop_Latino           VEP_pop_Male          VEP_pop_Total 
##                      0                      0                      0 
##          VEP_pop_White   Reported_votes_18_24   Reported_votes_25_34 
##                      0                      0                      0 
##   Reported_votes_35_44   Reported_votes_45_64 Reported_votes_65_plus 
##                      0                      0                      0 
##  Reported_votes_female  Reported_votes_latino    Reported_votes_male 
##                      0                      0                      0 
##   Reported_votes_white          VAP_pop_18_24          VAP_pop_25_34 
##                      0                      0                      0 
##          VAP_pop_35_44          VAP_pop_45_64        VAP_pop_65_plus 
##                      0                      0                      0 
##         VAP_pop_Female         VAP_pop_Latino           VAP_pop_Male 
##                      0                      0                      0 
##         VAP_pop_census          VAP_pop_White 
##                      0                      0

This chart shows that there are no missing values for any of the variables.

Part 2: Aggregate Turnout in the US States

Question 3

mydata$VEP.turnout <- c(mydata$Ballots_cast/mydata$VEP_pop)
print (mydata$VEP.turnout) #The voting eligible turnout rates from the administrative data

##  [1] 0.6177296 0.6806963 0.6795095 0.5626161 0.6752435 0.7669086 0.6930230
##  [8] 0.6946968 0.6539150 0.7184420 0.6789074 0.5518227 0.6682436 0.6595314
## [15] 0.6083741 0.7248151 0.6559589 0.6437515 0.6357299 0.7494134 0.6920059
## [22] 0.7048978 0.7300178 0.7937856 0.5986450 0.6597651 0.7203099 0.6877055
## [29] 0.6603633 0.7404666 0.7061192 0.6075056 0.6058103 0.7251225 0.6297917
## [36] 0.6645301 0.5516146 0.7497555 0.6966064 0.6256171 0.6540035 0.6562631
## [43] 0.5951490 0.6063745 0.6795739 0.7155281 0.7176365 0.7561560 0.5666433
## [50] 0.7503134 0.6493084

mydata$VAP.turnout <- c(mydata$Ballots_cast/mydata$VAP_pop)
print(mydata$VAP.turnout) #The voting age turnout rates from the administrative data.

##  [1] 0.5930843 0.6491820 0.6076447 0.5265955 0.5694937 0.7197238 0.6354156
##  [8] 0.6411542 0.6074273 0.6367952 0.6071643 0.4991455 0.6239323 0.6055255
## [15] 0.5833352 0.6896270 0.6160640 0.6128770 0.6029316 0.7363830 0.6311846
## [22] 0.6424129 0.7002890 0.7466368 0.5782625 0.6333446 0.7081734 0.6473529
## [29] 0.5802934 0.7162715 0.6256536 0.5634046 0.5388685 0.6753779 0.6088492
## [36] 0.6438075 0.5177236 0.7036222 0.6690085 0.5834684 0.6236186 0.6300328
## [43] 0.5657558 0.5162862 0.6355231 0.6990919 0.6609791 0.6773443 0.5554944
## [50] 0.7146876 0.6252005

mydata$VAP.census.turnout <-  c((mydata$Reported_votes_total*1000)/mydata$VAP_pop)
print(mydata$VAP.census.turnout) #The voting age turnout rates based on the US Census survey data.

##  [1] 0.5736111 0.5958615 0.6546384 0.5123108 0.5497127 0.6269171 0.5855097
##  [8] 0.6216455 0.7902503 0.5592658 0.5935685 0.5473953 0.6059521 0.6079597
## [15] 0.5773500 0.6599064 0.5816396 0.6338817 0.5728808 0.6883419 0.6579884
## [22] 0.5747641 0.6313509 0.7347507 0.6738138 0.6258176 0.7097672 0.6037736
## [29] 0.5578410 0.7091612 0.6378448 0.5719627 0.5383769 0.5843296 0.6276640
## [36] 0.6661799 0.5410442 0.7118248 0.6535989 0.5803615 0.6101383 0.5665106
## [43] 0.6198793 0.5417898 0.5918441 0.6507109 0.6334873 0.6386302 0.5403588
## [50] 0.7049272 0.6325082

According to the data, the voting age turnout rates based on the US Census survey data were the lowest and the actual turnout rates were higher with voting eligible turnout rates being the highest

Question 4

mydata$diff.VAP.VEP.admin <- c(mydata$VAP.turnout - mydata$VEP.turnout) #The difference between voting age and voting eligible turnout rates based on the administrative data
print(mydata$diff.VAP.VEP.admin)

##  [1] -0.02464535 -0.03151428 -0.07186480 -0.03602065 -0.10574982 -0.04718476
##  [7] -0.05760747 -0.05354257 -0.04648771 -0.08164681 -0.07174316 -0.05267723
## [13] -0.04431125 -0.05400597 -0.02503889 -0.03518811 -0.03989493 -0.03087449
## [19] -0.03279835 -0.01303041 -0.06082128 -0.06248494 -0.02972878 -0.04714883
## [25] -0.02038250 -0.02642051 -0.01213649 -0.04035262 -0.08006987 -0.02419515
## [31] -0.08046562 -0.04410097 -0.06694184 -0.04974460 -0.02094244 -0.02072264
## [37] -0.03389101 -0.04613335 -0.02759789 -0.04214872 -0.03038488 -0.02623028
## [43] -0.02939316 -0.09008836 -0.04405078 -0.01643627 -0.05665746 -0.07881170
## [49] -0.01114891 -0.03562584 -0.02410795

print(paste("Mean:" ,mean(mydata$diff.VAP.VEP.admin, na.rm = TRUE)))

## [1] "Mean: -0.0438273067203037"

print(paste("Range:" , range(mydata$diff.VAP.VEP.admin, na.rm = FALSE)))

## [1] "Range: -0.105749815454665"  "Range: -0.0111489086043757"

mydata$diff.VAP.admin.census <- c(mydata$VAP.turnout - mydata$VAP.census.turnout) #The difference between voting age turnout based on the administrative data and the self-reported Census VAP turnout
print(mydata$diff.VAP.admin.census)

##  [1]  0.0194731661  0.0533205735 -0.0469936689  0.0142846591  0.0197809688
##  [6]  0.0928067171  0.0499059040  0.0195087348 -0.1828229928  0.0775294262
## [11]  0.0135957308 -0.0482498555  0.0179801739 -0.0024342474  0.0059852462
## [16]  0.0297206655  0.0344243673 -0.0210047168  0.0300508379  0.0480410561
## [21] -0.0268037788  0.0676487956  0.0689381055  0.0118861016 -0.0955512884
## [26]  0.0075269740 -0.0015937606  0.0435793214  0.0224523781  0.0071102973
## [31] -0.0121912528 -0.0085580990  0.0004915996  0.0910483276 -0.0188147752
## [36] -0.0223724702 -0.0233206316 -0.0082025802  0.0154095916  0.0031069060
## [41]  0.0134803587  0.0635222348 -0.0541234664 -0.0255036343  0.0436790326
## [46]  0.0483809285  0.0274917206  0.0387140168  0.0151356384  0.0097604136
## [51] -0.0073077288

print(paste("Mean:" , mean(mydata$diff.VAP.admin.census, na.rm = TRUE)))

## [1] "Mean: 0.0101945494435088"

print(paste("Range:" , range(mydata$diff.VAP.admin.census, na.rm = FALSE)))

## [1] "Range: -0.182822992755451" "Range: 0.0928067170513283"

The mean and range for the difference in voting age turnout rates and voting eligible turnout rates are negative meaning that VEP turnout rates are higher than VAP rates and that the turnout among eligible voters is higher than the general voting age population. A possible explanation for this could be that those who are eligible to vote are more engaged and interested in voting.

The mean for the difference between voting age turnout and self reported Census data is positive meaning that there is higher voting age turnout compared to the Census data in most states. I say most as the range for the difference is from -0.18 to +0.09. A possible explanation for these results could be level of accuracy as administrative data may be more precise and frequent than census data.

Question 5

mydata$State[which.max(mydata$VAP.turnout)] #displays the state with the highest turnout rate

## [1] "MINNESOTA"

max(mydata$VAP.turnout, na.rm = TRUE) #displays the highest turnout rate.

## [1] 0.7466368

mydata$State[which.min(mydata$VAP.turnout)] #gives us the state with the lowest voter turnout rate

## [1] "HAWAII"

min(mydata$VAP.turnout, na.rm = TRUE) #displays the lowest turnout rate

## [1] 0.4991455

mydata$State[which.max(mydata$VAP.census.turnout)] #displays the state with the highest census turnout rate

## [1] "DISTRICT OF COLUMBIA"

max(mydata$VAP.census.turnout, na.rm = TRUE) #displays the highest census turnout rate.

## [1] 0.7902503

mydata$State[which.min(mydata$VAP.census.turnout)] #gives us the state with the lowest census turnout rate

## [1] "ARKANSAS"

min(mydata$VAP.census.turnout, na.rm = TRUE) #displays the lowest census turnout rate

## [1] 0.5123108

The state withe the highest census turnout rate is District of Colombia with a census turnout rate of 0.79 while the state with the lowest census turnout rate is Arkansas with a rate of 0.51. The state with the highest administrative data turnout rate is Minnesota with a rate of 0.74 and the lowest administrative data turnout rate is Hawaii with a rate of 0.49. While the are very close, the census survey has a larger range for it’s data than the administrative. I would say the administrative data is more accurate.

Part:3 Turnout by Age

Question 6

mydata$VAP_rate_18_24 <- mydata$Reported_votes_18_24/mydata$VAP_pop_18_24 #turnout rates for voting age adults age 18-24
print(paste("Mean of turnout for voting age adults 18-24:" , mean(mydata$VAP_rate_18_24, na.rm = TRUE)))

## [1] "Mean of turnout for voting age adults 18-24: 0.488670019433395"

mydata$VAP_Rate_25_34 <- mydata$Reported_votes_25_34/mydata$VAP_pop_25_34 
print(paste("Mean of turnout for voting age adults 25-34:" , mean(mydata$VAP_Rate_25_34, na.rm = TRUE)))

## [1] "Mean of turnout for voting age adults 25-34: 0.547481975299716"

mydata$VAP_Rate_35_44 <- mydata$Reported_votes_35_44/mydata$VAP_pop_35_44 
print(paste("Mean of turnout for voting age adults 35-44:" , mean(mydata$VAP_Rate_35_44, na.rm = TRUE)))

## [1] "Mean of turnout for voting age adults 35-44: 0.593840373296011"

mydata$VAP_Rate_45_64 <- mydata$Reported_votes_25_34/mydata$VAP_pop_45_64 
print(paste("Mean of turnout for voting age adults 45-64:" , mean(mydata$VAP_Rate_45_64, na.rm = TRUE)))

## [1] "Mean of turnout for voting age adults 45-64: 0.3036716395508"

mydata$VAP_Rate_65_plus <- mydata$Reported_votes_65_plus/mydata$VAP_pop_65_plus 
print(paste("Mean of turnout for voting age adults 65 plus:" , mean(mydata$VAP_Rate_65_plus, na.rm = TRUE)))

## [1] "Mean of turnout for voting age adults 65 plus: 0.734145682431118"

#Insert paragraph here

###Question 7

mydata$State[which.max(mydata$VAP_rate_18_24)] #gives us the state with the highest turnout rate for adults ages 18-24

## [1] "DISTRICT OF COLUMBIA"

max(mydata$VAP_rate_18_24) #displays the highest turnout rate for adults 18-24

## [1] 0.7460317

mydata$State[which.min(mydata$VAP_rate_18_24)] #gives us the state with the lowest turnout rate for adults ages 18-24

## [1] "OKLAHOMA"

min(mydata$VAP_rate_18_24) #displays the lowest turnout rate for adults 18-24

## [1] 0.2973761

Distict of columbia has the highest turnout rate among those 18-24 with a rate of 0.74 while Oklahoma has the lowest turnout rate among those 18-24 with a rate of 0.29

PA_turnout <- mydata$VAP_rate_18_24[mydata$State == "PENNSYLVANIA"] 
print(PA_turnout)

## [1] 0.4947569

FL_turnout <- mydata$VAP_rate_18_24[mydata$State == "FLORIDA"] 
print(FL_turnout)

## [1] 0.4164524

NY_turnout <- mydata$VAP_rate_18_24[mydata$State == "NEW YORK"] 
print(NY_turnout)

## [1] 0.432734

GA_turnout <- mydata$VAP_rate_18_24[mydata$State == "GEORGIA"] 
print(GA_turnout)

## [1] 0.4939394

CA_turnout <- mydata$VAP_rate_18_24[mydata$State == "CALIFORNIA"] 
print(CA_turnout)

## [1] 0.4846676

# This code was to find the turnout rates for ages 18-24 in the states of my choice

youth_turnout <- c(0.4947569, 0.4164524, 0.432734, 0.4939394, 0.4846676)
names(youth_turnout) <- c("PA","FL","NY","GA","CA")
print(youth_turnout)

##        PA        FL        NY        GA        CA 
## 0.4947569 0.4164524 0.4327340 0.4939394 0.4846676

The highest turnout rate is in PA with a rate of about 0.494 and the lowest turnout rate is FL with a turnout rate of 0.416. The difference between the rates of these two state is about 0.078. Looking back at the data, the national mean of voter turnout for ages 18-24 is 0.488. This shows that PA is a little bit higher than the national average along with GA. FL and NY are lower than the national average and CA is almost at the national average. I think the fact that these are all states with big cities and high populations are why their turnout rates are close to the national average. I think florida has the lowest rate due to it having a higher elderly population and that there isnt much of an effort to promote youth voting as in other states such as PA.

Question 8

VEP turnout measured voter eligibility meaning those who had the correct citizen and legal status to vote while VAP turnout only measured those who were old enough to vote. the difference in data being that voter eligibility(VEP) was normally smaller than VAP in states. VAP census measured the number of those who were of voting age as of the last census. I think the most accurate of the variable would be the voter eligibility turnout (VEP) as it measured not only age, but took into account other factors that may prevent people from voting. Both VAP and VAP census were too broad and didn’t account for other factor or were out dated.

Looking at the turnout rates for each group I noticed that it got higher up until ages 34-44 then it is the highest at ages 65 plus. The group that ,in my opinion, is the most capable of voting ,ages 18-24, come in second to last. I believe that many individuals in this age group do not feel the urge and importance to vote as much as other age groups. Maybe it has to do with generational differences.

due to majority of the voters beign on the elderly side a lot of policy making will reflect that age groups views and mindset. While the younger voters don’t get their voices reflected much in policy making. A strategy that could be implemented to address the under representation of younger voters would be to better inform people about the voting process befor ethey reach voting age so that when they do its not as hard.

Looking at the data I realized that not enough people vote and those who will be members of this coutnry for much longer than others do not vote nearly as much as those who, to put it bluntly, will not be here to live through the consequences of their decisions. This shows that the health of american dmeocracy is low.