Data Analytics and Computational Social Science: HNguyen HW 4

Henry Nguyen

Load Packages

library(tidyverse)
library(dplyr)
library(ggplot2)

HIV

1a. Readin your dataset

Import Data HIV Incidence Data

# Data set #1
# Data obtained from here:https://gis.cdc.gov/grasp/nchhstpatlas/tables.html

HIV.State <- read_csv("HIV.by.State.CSV", skip = 10)

HIV.State <- select(HIV.State, 2,3,5,6)

HIV.State <- rename(HIV.State, Rate =4)

HIV.State.Wide.Year <- HIV.State%>%
  select(1:3)%>%
  pivot_wider(
    names_from = Year,
    values_from = Cases
  )

1b - Compute descriptive Satistics

I will compute the mean, median, and standard deviation by case-state by year and total year.

Here I calculate the mean and median for cases across all states from 2008-2021.

HIV.Statistics.Total <- HIV.State%>%
  summarise(
    HIV.Mean.Total = mean(Cases),
    HIV.Median.Total = median(Cases))

HIV.Statistics.Total

# A tibble: 1 × 2
  HIV.Mean.Total HIV.Median.Total
           <dbl>            <dbl>
1           754.             352.

Next, I calculate the mean,median, and standard deviation for each year from 2008-2019.

Rate calculations are from 2008-2019 as there is not data for the years 2020 and 2021.

Now I calculate the mean, median, and SD for each state from 2008-2021

By.State <- group_by(HIV.State, Geography)



HIV.Statistics.by.State <- summarise(By.State,
                  HIV.Mean.by.State = mean(Cases),
                  HIV.Median.by.State = median(Cases),
                  HIV.SD.by.State = sd(Cases),
                  )

HIV.Statistics.by.State <- arrange(HIV.Statistics.by.State, HIV.Median.by.State)

HIV.Statistics.by.State

# A tibble: 51 × 4
   Geography     HIV.Mean.by.State HIV.Median.by.State HIV.SD.by.State
   <chr>                     <dbl>               <dbl>           <dbl>
 1 Vermont                    13.8                14              5.10
 2 Wyoming                    14.2                14.5            5.51
 3 North Dakota               23.6                19.5           12.6 
 4 Montana                    20.9                20.5            6.07
 5 Alaska                     28                  27.5            7.57
 6 South Dakota               29.1                29.5            7.29
 7 Idaho                      36.6                36.5            9.39
 8 New Hampshire              36.6                38              8.08
 9 Maine                      41.4                46             13.6 
10 Rhode Island               82.2                77.5           25.2 
# … with 41 more rows

Rate.By.State <- filter(By.State, Year < 2020)

Rate.By.State$Rate = as.numeric(Rate.By.State$Rate)

Rate.By.State <- group_by(Rate.By.State, Geography)

HIV.Statistics.by.State.Rate <- summarise(Rate.By.State,
                  HIV.Rate.Mean.by.State = mean(Rate),
                  HIV.Rate.Median.by.State = median(Rate),
                  HIV.Rate.SD.by.State = sd(Rate),
                  )

HIV.Statistics.by.State.Rate <- arrange(HIV.Statistics.by.State.Rate, HIV.Rate.Median.by.State)

HIV.Statistics.by.State.Rate

# A tibble: 51 × 4
   Geography     HIV.Rate.Mean.by… HIV.Rate.Median.b… HIV.Rate.SD.by.…
   <chr>                     <dbl>              <dbl>            <dbl>
 1 Montana                    2.62               2.55            0.589
 2 Vermont                    2.78               2.85            0.805
 3 Idaho                      2.91               2.9             0.779
 4 North Dakota               3.86               3.2             1.99 
 5 Wyoming                    3.24               3.25            1.10 
 6 New Hampshire              3.37               3.4             0.656
 7 Maine                      3.92               4.2             0.972
 8 South Dakota               4.29               4.25            0.908
 9 Iowa                       4.45               4.55            0.485
10 Alaska                     4.93               4.6             1.19 
# … with 41 more rows

Healthcare GDP

Import and tidy Healthcare GDP data

I tidy the data even more. Ambulatory Health services and hospitals are including in health and social assistance along with other industries, therefore I will remove them.

GDP.State <- read_csv("StateGDP.csv", skip = 4)

GDP.State <- GDP.State %>% 
  select(2,4:17)

GDP.State <- GDP.State %>%
  pivot_longer(
    `2008`: `2020`,
    names_to = "Year",
    values_to = "GDP"
  )

GDP.State <- filter(GDP.State,Description == "Health care and social assistance")

1b cont. GDP Statistics total

GDP.Statistics.Total <- summarise(GDP.State,
  GDP.Mean.Total = mean(GDP),
  GDP.Median.Total = median(GDP)
)

GDP.Statistics.Total

# A tibble: 1 × 2
  GDP.Mean.Total GDP.Median.Total
           <dbl>            <dbl>
1         65088.           21526.

GDP statitistics by Year

GDP.by.YEAR <- group_by(GDP.State, Year)

GDP.Statistics.by.Year <- summarise(GDP.by.YEAR,
            GDP.Mean.Year = mean(GDP),
            GDP.Median.Year = median(GDP),
            GDP.SD.Year = sd(GDP)
    
  )

GDP.by.YEAR

# A tibble: 780 × 4
# Groups:   Year [13]
   GeoName         Description                       Year      GDP
   <chr>           <chr>                             <chr>   <dbl>
 1 United States * Health care and social assistance 2008  1017197
 2 United States * Health care and social assistance 2009  1078771
 3 United States * Health care and social assistance 2010  1112327
 4 United States * Health care and social assistance 2011  1149944
 5 United States * Health care and social assistance 2012  1195074
 6 United States * Health care and social assistance 2013  1230767
 7 United States * Health care and social assistance 2014  1266432
 8 United States * Health care and social assistance 2015  1337785
 9 United States * Health care and social assistance 2016  1406434
10 United States * Health care and social assistance 2017  1461525
# … with 770 more rows

GDP statistics by State

GDP.by.State <- group_by(GDP.State, GeoName)

GDP.Statistics.by.State <- summarise(GDP.by.State,
                                     GDP.Mean.State = mean(GDP),
                                     GDP.Median.State = median(GDP),
                                     GDP.SD.State = sd(GDP)
                                       )

GDP.Statistics.by.State <- arrange(GDP.Statistics.by.State, GDP.Median.State)
GDP.Statistics.by.State

# A tibble: 60 × 4
   GeoName      GDP.Mean.State GDP.Median.State GDP.SD.State
   <chr>                 <dbl>            <dbl>        <dbl>
 1 Wyoming               1584.            1586.         131.
 2 Vermont               3172.            3091.         412.
 3 Alaska                3542.            3474.         633.
 4 North Dakota          3664.            3603.         763.
 5 Montana               4290.            4116.         749.
 6 South Dakota          4251.            4125.         793.
 7 Delaware              4725.            4704.         738.
 8 Hawaii                5272.            5077.         868.
 9 Idaho                 5333.            5147.        1036.
10 Rhode Island          5571.            5557.         464.
# … with 50 more rows

2 Create Visulizations

HIV

This goal of this visualization is to see if there are states with a different trend in incident HIV infection and rate.

#Figure 1a: Trend in Cases From 2008-2021 by ACross the States of the United States
F1A.point.hiv.ALL.cases <- ggplot(data=HIV.State, mapping = aes(x= Year, y= Cases, group = Geography))+
  geom_line()


F1A.point.hiv.ALL.cases

#Figure 1b: Trend in Cases From 2008-2021 by Individual States
F1b.point.hiv.ALL.cases.WRAPPED <- ggplot(data=HIV.State, mapping = aes(x= Year, y= Cases, group = Geography))+
  geom_line()+
  facet_wrap(~Geography, nrow=12,, scales = "free_y")


F1b.point.hiv.ALL.cases.WRAPPED

#Figure 1c: Trend in Cases From 2008-2021 by Individual States
F1C.point.hiv <- ggplot(data=HIV.State, mapping = aes(x= Year, y= Cases))+
  geom_point(position = "jitter")+
  geom_smooth()+
  facet_wrap(~Geography, nrow=12, scales = "free")

F1C.point.hiv

#Figure 2a: Trend in Incident Rate of HIV Infection From 2008-2019 Across the States of the United States

F2a.point.Rate.hiv.LINE <- ggplot(data = Rate.By.State, mapping = aes(x=Year,y=Rate, group=Geography))+
  geom_point(position = "jitter")+
  geom_line()

F2a.point.Rate.hiv.LINE

#Figure 2b: Trend in Incident Rate of HIV Infection From 2008-2019 by state
   
F2b.point.Rate.hiv.LINE.WRAP <- ggplot(data = Rate.By.State, mapping = aes(x=Year,y=Rate, group=Geography))+
  geom_point()+
  geom_line()+
  facet_wrap(~Geography, nrow =12, scales = "free")+
  geom_smooth()
  
F2b.point.Rate.hiv.LINE.WRAP

Healthcare GDP

The goal of this visualization is to see the change in healthcare gdp over time by state

# Figure 3a. Trend in Healthcare GDP across the United States from 2008-2021.
F3a.point.gdp <- ggplot(data = GDP.State, mapping = aes(x=Year, y=GDP, group = GeoName))+
  geom_point()+
  geom_line()

F3a.point.gdp

# Figure 3b. Trend in Health care GDP across the United States by state from 2008-2021.
F3b.point.gdp.WRAPPED <- ggplot(data = GDP.State, mapping = aes(x=Year, y=GDP, group = GeoName))+
  geom_point()+
  geom_smooth()+
  facet_wrap(~GeoName,, nrow =12, scales = "free")

F3b.point.gdp.WRAPPED

3. Describe Visulations

3a. What variables are being visualized?

These visualization looks at the following variables:

Trends in incident HIV infections across the United States from 2008-2021 (Figure 1a).
Trends in incident HIV infections by state from 2008-2021 (Figure 1b and 1c).
Trends in rate of incident HIV infections from 2008-2019 across the United States (Figure 2a).
Trends in rate of incident HIV infections from 2008-2019 by state (Figure 2b).

These visualizations look at the following variables:

Trend in healthcare GDP across the United States from 2008-2021 (Figure 3a).
Trend in health care GDP by state from 2008-2021 (Figure 3b).

3b. What questions are you answering with the visualizations?

Figures 1a and 3a seeks to answer the questions ” What are the trends in incident HIV infections, rate of HIV infections, and healthcare GDP across the United States over this time frame?”
Figure 1b,1c,2b, and 3b seeks to answer the questions ” What are the trends in incident HIV infections, rate of HIV infections, and healthcare GDP by state over this time frame?”

3c. What conclusions can you make from the visualization?

There is an overall trend that incident HIV infections have decreased from 2008-2021 (Figure 1a and 2a).
Incident cases have decreased in all states with some states having outlier years where incident cases increased in recent years. For example North Dakota, Tennessee, Utah, Washington, West Virginia, and Wyoming (Figure 1b and 1c).
Looking at the incidence rates over time, you can see that most states had case rates decrease over time (Figure 2b).
Looking at rates, some states appear to have a plateau in case rates from an increasing trend ( i.e Arizona, Arkansas, Nevada, North Dakota) (Figure 2b).
Looking at rates, some states appear to have increase case rates (i.e Missisippi, Montana, New Mexico, North Dakota, Ohio, Oklahoma, Washington, West Virginia) (Figure 2b).
Looking at healthcare GDP, most states have increased their healthcare GDP over time (Figure 3a and 3b).

4 Indentify Limitations to Visualizations

There are major differences in cases and rates by size of states population. Using a “free scale” is helpful for visualization but may limit accuracy. This also applies to healthcare GDP.
There is case data from 2008-2021, but only rate data from 2008-2019.
The relationship between incident HIV cases/rates and healthcare GDP is still unclear. Although likely to be insignificant.
Without a titles, the visualization is probably unclear to the naive viewer.
Visualizations could be neater by commiting to a scale or moving the decimal so that visualization may be clearn (i.e healthcare GDP dollars).

Comment on this article Share:

HNguyen HW 4

Load Packages

HIV

1a. Readin your dataset

Import Data HIV Incidence Data

1b - Compute descriptive Satistics

Here I calculate the mean and median for cases across all states from 2008-2021.

Next, I calculate the mean,median, and standard deviation for each year from 2008-2019.

Now I calculate the mean, median, and SD for each state from 2008-2021

Healthcare GDP

Import and tidy Healthcare GDP data

1b cont. GDP Statistics total

GDP statitistics by Year

GDP statistics by State

2 Create Visulizations

HIV

Healthcare GDP

3. Describe Visulations

3a. What variables are being visualized?

3b. What questions are you answering with the visualizations?

3c. What conclusions can you make from the visualization?

4 Indentify Limitations to Visualizations

Reuse

Citation