library(rvest)
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
surveyData <- list()
#Defines URL, I made this easier to loop by using paste0 to add the ID
#to the generic URL
for (i in 1:8) {
j=i+31280769
namePage <- paste0("https://www.ncbi.nlm.nih.gov/biosample/?term=",j)
#testPage variable will store all data from the webpage
#similar to what you see with "Inspect Element"
testPage <- read_html(namePage)
#Since the data we want is in a table, this is a good first step
tableText <- testPage %>%
html_node("table") %>%
html_table()
names(tableText)<-c('Question','Response')
surveyData[[as.character(j)]] <-tableText
}
head(tableText)
## # A tibble: 6 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 21.4
## 6 host body product UBERON:feces
surveyData
## $`31280770`
## # A tibble: 230 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 21.0
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 162.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 220 more rows
##
## $`31280771`
## # A tibble: 295 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 26.0
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 184.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 285 more rows
##
## $`31280772`
## # A tibble: 293 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am left handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 25.7
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 159.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 283 more rows
##
## $`31280773`
## # A tibble: 295 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am left handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 25.4
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 166.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 285 more rows
##
## $`31280774`
## # A tibble: 230 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 20.4
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 158.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 220 more rows
##
## $`31280775`
## # A tibble: 293 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 33.3
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 154.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 283 more rows
##
## $`31280776`
## # A tibble: 292 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 28.9
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 159.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 282 more rows
##
## $`31280777`
## # A tibble: 293 × 2
## Question Response
## <chr> <chr>
## 1 dominant hand I am right handed
## 2 environmental medium feces
## 3 environmental package human-gut
## 4 host body habitat UBERON:feces
## 5 host body mass index 21.4
## 6 host body product UBERON:feces
## 7 host tissue sampled UBERON:feces
## 8 host height 156.0
## 9 life stage Adult
## 10 race Hispanic
## # ℹ 283 more rows
responses <- sapply(surveyData, function(x) x[2])
bmi <- sapply(responses, function(x) x[5])
pie(table(bmi), main = "Pie Chart of Participant BMI")
alcohol_types_beercider <- sapply(responses, function(x) x[31])
pie(table(alcohol_types_beercider), main = "Pie Chart of Beer Consumption")
