Skills Drill 2

Data Prep

library(readr)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(ggplot2)

## Warning in file(con, "r"): cannot open file '/var/db/timezone/zoneinfo/
## +VERSION': No such file or directory

Abbreviated_Voter_Dataset_Labeled <- read_csv("/Users/chelsyrodriguez/Downloads/Abbreviated Voter Dataset Labeled.csv")

## 
## ── Column specification ────────────────────────────────────────────────────────
## cols(
##   .default = col_character(),
##   NumChildren = col_double(),
##   Immigr_Economy_GiveTake = col_double(),
##   ft_fem_2017 = col_double(),
##   ft_immig_2017 = col_double(),
##   ft_police_2017 = col_double(),
##   ft_dem_2017 = col_double(),
##   ft_rep_2017 = col_double(),
##   ft_evang_2017 = col_double(),
##   ft_muslim_2017 = col_double(),
##   ft_jew_2017 = col_double(),
##   ft_christ_2017 = col_double(),
##   ft_gays_2017 = col_double(),
##   ft_unions_2017 = col_double(),
##   ft_altright_2017 = col_double(),
##   ft_black_2017 = col_double(),
##   ft_white_2017 = col_double(),
##   ft_hisp_2017 = col_double()
## )
## ℹ Use `spec()` for the full column specifications.

Response summary

• I hypothesize there is a relationship between areatype and GunOwnership. I think this is important to study because we want to identify if respondents are capable or are even allowed to own a gun depending on where they currently reside in. To be specific, my analysis will investigate if the area the respondent resides in (DV) is dependent upon owning a gun in their household (IV).

• IV Response Summary

table(Abbreviated_Voter_Dataset_Labeled$GunOwnership) %>%
prop.table()

## 
##           Dont Know    Gun in Household No Gun in Household 
##          0.06677648          0.39039534          0.54282818

*DV Response Summary

table(Abbreviated_Voter_Dataset_Labeled$areatype) %>%
prop.table()

## 
##        City       Other  Rural Area      Suburb        Town 
## 0.284652218 0.006426411 0.186869960 0.378276210 0.143775202

Primary Investigation

*Expected values

chisq.test(Abbreviated_Voter_Dataset_Labeled$areatype, Abbreviated_Voter_Dataset_Labeled$GunOwnership) [7]

## Warning in chisq.test(Abbreviated_Voter_Dataset_Labeled$areatype,
## Abbreviated_Voter_Dataset_Labeled$GunOwnership): Chi-squared approximation may
## be incorrect

## $expected
##             
##              Dont Know Gun in Household No Gun in Household
##   City       149.60910        872.52938          1212.86152
##   Other        3.34696         19.51967            27.13337
##   Rural Area  97.32959        567.63209           789.03832
##   Suburb     198.67552       1158.68779          1610.63669
##   Town        75.03883        437.63107           608.33010

*Observed Values

chisq.test(Abbreviated_Voter_Dataset_Labeled$areatype, Abbreviated_Voter_Dataset_Labeled$GunOwnership) [6]

## Warning in chisq.test(Abbreviated_Voter_Dataset_Labeled$areatype,
## Abbreviated_Voter_Dataset_Labeled$GunOwnership): Chi-squared approximation may
## be incorrect

## $observed
##             
##              Dont Know Gun in Household No Gun in Household
##   City             144              684                1407
##   Other              0               21                  29
##   Rural Area       122              841                 491
##   Suburb           173             1053                1742
##   Town              85              457                 579

I’ll be comparing respondents that live in the city and respondents that live in the suburb. I expected 873 respondents that live in the city own a gun in their household, however I observed 684 respondents that reside in the city own a gun. *I expected 1,159 respondents that live in the suburbs own a gun in their household, howeverI observed 1,053 respondents that reside in the suburbs own a gun.

Abbreviated_Voter_Dataset_Labeled %>%
  group_by(areatype,GunOwnership) %>%
  summarize(n=n()) %>%
  mutate(percent=n/sum(n)) %>%
  ggplot()+
  geom_col(aes(x=areatype,y=percent,fill=GunOwnership))

## `summarise()` has grouped output by 'areatype'. You can override using the `.groups` argument.

Chi-Square Test

chisq.test(Abbreviated_Voter_Dataset_Labeled$areatype, Abbreviated_Voter_Dataset_Labeled$GunOwnership)

## Warning in chisq.test(Abbreviated_Voter_Dataset_Labeled$areatype,
## Abbreviated_Voter_Dataset_Labeled$GunOwnership): Chi-squared approximation may
## be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  Abbreviated_Voter_Dataset_Labeled$areatype and Abbreviated_Voter_Dataset_Labeled$GunOwnership
## X-squared = 353.36, df = 8, p-value < 2.2e-16

According to the results of the chi-square test, the p-value identified the scientific notation as < 2.2e-16 which is a smaller value than .05. There is a statistically significant relationship between areatype and GunOwnership. This results reject the null hypothesis.