DATA_607_Week5_Homework

DATA 607 Week5 Homework

Let’s load the required libraries in R for data analysis

library(dplyr)
library(downloader)
library(stringr)
library(htmlTable)
library(tidyverse)
library(flextable)

Download and read the CSV file to a dataframe object

View the data

# Import the data from the csv file
vaccination_data <- read.csv("https://raw.githubusercontent.com/baruab/msdsrepo/main/DATA-607/israeli_vaccination_data.csv")

vaccination_data  %>% regulartable() %>% autofit()  

Seperate the data by even and odd rows

# Extract Odd rows from dataframe
odd_rows <- seq_len(nrow(vaccination_data)) %% 2
# Drop even rows
data_odd_rows <- vaccination_data[odd_rows == 1, ]

# Now Drop odd rows, create a dataframe of even rows
data_even_rows <- vaccination_data[odd_rows == 0, ]

subset_even_rows <- subset(data_even_rows, select= c(2, 3))

Rename the columns

Add new columns, converting the existing ones from character to numeric for calculations

View Population as numeric

# Rename even row column names
colnames(subset_even_rows) <- c("%_Not_Vax","%_Fully_Vax")


# Add numeric columns for the Vaccination percentage values
num_subset_even_rows <- subset_even_rows %>% 
  mutate(Percent_NotVax =  str_extract_all(subset_even_rows$`%_Not_Vax`, "^[:digit:]+.[:digit:]+")) %>%
  unnest() %>%
  mutate(Percent_NotVax = as.numeric(Percent_NotVax)) %>%
  
  mutate(Percent_FullyVax =  str_extract_all(subset_even_rows$`%_Fully_Vax`, "^[:digit:]+.[:digit:]+")) %>%
  unnest() %>%
  mutate(Percent_FullyVax = as.numeric(Percent_FullyVax))


# Rename odd row column names
colnames(data_odd_rows) <- c("Age" ,"Population_Not_Vax","Population_Fully_Vax", "Severe_Not_Vax_Cases", "Severe_Fully_Vax_Cases", "Efficacy_vs_Severe_Disease")


# Add numeric columns for the Vaccination population values
num_data_odd_rows <- data_odd_rows %>% 
  mutate(Population_NotVax = as.numeric(gsub(",","", data_odd_rows$Population_Not_Vax ))) %>%
  unnest()  %>% 
  mutate(Population_FullyVax = as.numeric(gsub(",","", data_odd_rows$Population_Fully_Vax ))) %>%
  unnest() %>%

# Compute Efficacy vs Severe Disease     format(round(sum/count, 2), nsmall = 2) 
  mutate(Efficacy_vs_Severe_Disease = format(round(1 - ( data_odd_rows$Severe_Fully_Vax_Cases /       data_odd_rows$Severe_Not_Vax_Cases),2), nsmall = 2 )) %>%
  unnest()

num_data_odd_rows$Population_NotVax

## [1] 1116834  186078

num_data_odd_rows$Population_FullyVax  

## [1] 3501118 2133516

Combine the datasets with numeric columns for calculations

Add a Total row, sum the relevant columns

# Combine the two data frames with the new numeric columns created
data_rows <- cbind(subset(num_data_odd_rows, select = c(1,4,5,6,7,8)), subset(num_subset_even_rows, select= c(3, 4)))

# Add new columns totalling the percentage and population by row 
new_data_rows <- mutate(data_rows, total_percentage = data_rows$Percent_NotVax + data_rows$Percent_FullyVax ) %>%
                 mutate(data_rows, total_population = data_rows$Population_NotVax + data_rows$Population_FullyVax )  

new_data_rows[nrow(new_data_rows)+1, ] <- c('Total', sum(new_data_rows$Severe_Not_Vax_Cases), sum(new_data_rows$Severe_Fully_Vax_Cases),NA, sum(new_data_rows$Population_NotVax), sum(new_data_rows$Population_FullyVax),NA, NA, NA,  sum(new_data_rows$total_population) )

new_data_rows  %>% regulartable() %>% autofit()  

Display the final dataset

# Short Column names
colnames(new_data_rows) <- c("Age" ,"NotVax_Cases","FullyVax_Cases", "Effi_Ratio", "NotVax_Popu", "FullyVax_Popu", "NotVax_Perc", "FullyVax_Perc", "Total_Perc", "Total_Popu")

col_order <- c("Age", "NotVax_Popu", "FullyVax_Popu",  "Total_Popu", "NotVax_Perc", "FullyVax_Perc", "Total_Perc","NotVax_Cases","FullyVax_Cases", "Effi_Ratio")
final_data_rows <- new_data_rows[, col_order]

htmlTable(final_data_rows)

	Age	NotVax_Popu	FullyVax_Popu	Total_Popu	NotVax_Perc	FullyVax_Perc	Total_Perc	NotVax_Cases	FullyVax_Cases	Effi_Ratio
1	<50	1116834	3501118	4617952	23.3	73	96.3	43	11	0.74
2	50	186078	2133516	2319594	7.9	90.4	98.3	171	290	-0.70
3	Total	1302912	5634634	6937546				214	301

According to Wikipedia.org, Israel has an estimated population of 9.4 million.

Ref: https://en.wikipedia.org/wiki/Israel

Please note: Based on the news article below, Israelis over 16 years are only eligible for Vaccination

Ref: https://www.usnews.com/news/health-news/articles/2021-02-04/all-israelis-over-16-are-eligible-for-coronavirus-vaccine Age Structure Ref: https://www.statista.com/statistics/526596/age-structure-in-israel/

The total population in this study is approx 6.9 million, as per calculation in the Total row. Also note that the total percentage in each category is less than 100%. My assumption is the use case did not include the population of 16 years and under in this study. Based on the age structure reference link above, approx 27% of Israel’s population is under 15, which if included adds up to the estimated 9.4 million total population.

1) Based on the information provided in this use-case, we do not have enough details to calculate the total population of Israel. The first age range should have been ‘16 - 50’ instead of ‘<50’. This total population only represents only people greater than 16 yrs.

efficacy_data_rows <- subset(final_data_rows, select= c("Age","NotVax_Popu", "FullyVax_Popu",  "Total_Popu",
                                "NotVax_Cases","FullyVax_Cases", "Effi_Ratio" ))

# Add Efficacy ratio values
more_efficacy_ratio <- efficacy_data_rows %>% 
  mutate(Severecase_rate_NotVax = format(round( as.numeric(efficacy_data_rows$NotVax_Cases) / as.numeric(efficacy_data_rows$NotVax_Popu),8), nsmall =8 ))  %>%
  unnest() %>% 
  mutate( Severecase_rate_FullVax = format(round( as.numeric(efficacy_data_rows$FullyVax_Cases) / as.numeric(efficacy_data_rows$FullyVax_Popu),8), nsmall = 8 )) %>%
  unnest()

htmlTable(more_efficacy_ratio)

	Age	NotVax_Popu	FullyVax_Popu	Total_Popu	NotVax_Cases	FullyVax_Cases	Effi_Ratio	Severecase_rate_NotVax	Severecase_rate_FullVax
1	<50	1116834	3501118	4617952	43	11	0.74	0.00003850	0.00000314
2	50	186078	2133516	2319594	171	290	-0.70	0.00091897	0.00013593
3	Total	1302912	5634634	6937546	214	301		0.00016425	0.00005342

2) Above are the calculated values for Efficacy vs Disease, one of the value is positive and other is negative. It tells that the Severe cases for fully vaccinated people was higher in age group greater than 50. Clearly the cases were much higher in >50 age group. Also I have calculated the rate of ‘Number of Severe cases’ by their respective population. The ratio is fairly low indicating high efficacy rate of the vaccine is general.

3) There is no comparison with the efficacy vs severe disease with the rate of severe cases in individuals.