Boris Bikes
Keren Afriyie
2022-08-01
#Loading required packages
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.3.6 ✔ purrr 0.3.4
## ✔ tibble 3.1.8 ✔ dplyr 1.0.9
## ✔ tidyr 1.2.0 ✔ stringr 1.4.0
## ✔ readr 2.1.2 ✔ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(httr)
library(jsonlite)##
## Attaching package: 'jsonlite'
##
## The following object is masked from 'package:purrr':
##
## flattenlibrary(leaflet)
```r
#Downloading the bike point data
BorisBikesAPI <- 'https://api.tfl.gov.uk/BikePoint/'
get_BorisBikesAPI <- GET(BorisBikesAPI)
get_BorisBikes_text <- content(get_BorisBikesAPI, "text")
get_BorisBikesAPI_JSON <- fromJSON(get_BorisBikes_text, flatten = TRUE)
df_BorisBikes <- as.data.frame(get_BorisBikesAPI_JSON)head(df_BorisBikes, 2)## $type
## 1 Tfl.Api.Presentation.Entities.Place, Tfl.Api.Presentation.Entities
## 2 Tfl.Api.Presentation.Entities.Place, Tfl.Api.Presentation.Entities
## id url commonName placeType
## 1 BikePoints_84 /Place/BikePoints_84 Breams Buildings, Holborn BikePoint
## 2 BikePoints_85 /Place/BikePoints_85 Tanner Street, Bermondsey BikePoint
## additionalProperties
## 1 Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Description, Description, Description, Description, Description, Description, Description, Description, Description, Description, Description, TerminalName, Installed, Locked, InstallDate, RemovalDate, Temporary, NbBikes, NbEmptyDocks, NbDocks, NbStandardBikes, NbEBikes, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, 003449, true, false, 1279020240000, , false, 24, 1, 25, , , 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z, 2022-08-02T11:24:35.73Z
## 2 Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Tfl.Api.Presentation.Entities.AdditionalProperties, Tfl.Api.Presentation.Entities, Description, Description, Description, Description, Description, Description, Description, Description, Description, Description, Description, TerminalName, Installed, Locked, InstallDate, RemovalDate, Temporary, NbBikes, NbEmptyDocks, NbDocks, NbStandardBikes, NbEBikes, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, BikePoints, 000994, true, false, 1279026060000, , false, 33, 5, 41, , , 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z, 2022-08-02T10:53:18.247Z
## children childrenUrls lat lon
## 1 NULL NULL 51.51594 -0.111778
## 2 NULL NULL 51.50065 -0.078600Plotting the location of the bike points using leaflet
Below I use the package leaflet to plot the location of the bike points using the coordinates in the data frame.
#Subsetting the necessary data
df_key_BorisBikes <- df_BorisBikes[,c(2,4,5,9,10)]
head(df_key_BorisBikes)## id commonName placeType lat
## 1 BikePoints_84 Breams Buildings, Holborn BikePoint 51.51594
## 2 BikePoints_85 Tanner Street, Bermondsey BikePoint 51.50065
## 3 BikePoints_86 Sancroft Street, Vauxhall BikePoint 51.48948
## 4 BikePoints_87 Devonshire Square, Liverpool Street BikePoint 51.51647
## 5 BikePoints_88 Bayley Street , Bloomsbury BikePoint 51.51859
## 6 BikePoints_89 Tavistock Place, Bloomsbury BikePoint 51.52625
## lon
## 1 -0.111778
## 2 -0.078600
## 3 -0.115156
## 4 -0.079684
## 5 -0.132053
## 6 -0.123509#subsetting the data again to generate a dataframe of the coordinates
coordinates <- df_key_BorisBikes[, c(2,4,5)]
head(coordinates)## commonName lat lon
## 1 Breams Buildings, Holborn 51.51594 -0.111778
## 2 Tanner Street, Bermondsey 51.50065 -0.078600
## 3 Sancroft Street, Vauxhall 51.48948 -0.115156
## 4 Devonshire Square, Liverpool Street 51.51647 -0.079684
## 5 Bayley Street , Bloomsbury 51.51859 -0.132053
## 6 Tavistock Place, Bloomsbury 51.52625 -0.123509#plotting the coordinates on a map
map <- leaflet () %>%
addTiles() %>%
addMarkers(lng = coordinates[,'lon'], lat = coordinates[,'lat'], popup = coordinates[,'commonName'])mapWhere are bike points most frequent?
Below I extract the name of the bike point after the comma to try and determine where bike points are most frequent. This was not a perfect method as many of the locations overlap. If I had more time I would perform a more accurate measurement gathering.
In addition, I would have compared the number of bike points the population of the area to generate a per capita metric.
names <- sub(".*,", "", df_key_BorisBikes$commonName)
unique_names <- unique(names)#Adding the area names to the dataframe
df_key_BorisBikes$area <- names
head(df_key_BorisBikes)## id commonName placeType lat
## 1 BikePoints_84 Breams Buildings, Holborn BikePoint 51.51594
## 2 BikePoints_85 Tanner Street, Bermondsey BikePoint 51.50065
## 3 BikePoints_86 Sancroft Street, Vauxhall BikePoint 51.48948
## 4 BikePoints_87 Devonshire Square, Liverpool Street BikePoint 51.51647
## 5 BikePoints_88 Bayley Street , Bloomsbury BikePoint 51.51859
## 6 BikePoints_89 Tavistock Place, Bloomsbury BikePoint 51.52625
## lon area
## 1 -0.111778 Holborn
## 2 -0.078600 Bermondsey
## 3 -0.115156 Vauxhall
## 4 -0.079684 Liverpool Street
## 5 -0.132053 Bloomsbury
## 6 -0.123509 Bloomsburyarea_freq <- df_key_BorisBikes %>%
count(area, sort = TRUE)
head(area_freq)## area n
## 1 Marylebone 26
## 2 Holborn 20
## 3 Mayfair 17
## 4 Westminster 16
## 5 Kensington 15
## 6 Vauxhall 15#Areas the have the most bike points
top_10_areas <- area_freq[1:10,]
top_10_areas## area n
## 1 Marylebone 26
## 2 Holborn 20
## 3 Mayfair 17
## 4 Westminster 16
## 5 Kensington 15
## 6 Vauxhall 15
## 7 Battersea 14
## 8 Bethnal Green 14
## 9 Bloomsbury 14
## 10 Hoxton 13nrow(area_freq)## [1] 123#Areas with the least amount of bike points
lowest_10_areas <- area_freq[114:123,]
lowest_10_areas## area n
## 114 Parson's Green 1
## 115 Shad Thames 1
## 116 Sloane Square 1
## 117 St Lukes 1
## 118 St Pauls 1
## 119 St.John's Wood 1
## 120 Stratford 1
## 121 Wandsworth Common 1
## 122 Hoxton 1
## 123 Shepherd's Bush 1top_10_areas %>%
ggplot(aes(x = area, y = n)) +
geom_bar(stat = 'identity', fill = 'green') +
theme(axis.text.x = element_text(angle=45, hjust = 1))+
labs(x = "Area", y = "# of Bike Points by Area",
title = "Areas that have the Greatest # of Bike Points")
lowest_10_areas %>%
ggplot(aes(x = area, y = n)) +
geom_bar(stat = 'identity', fill = 'red') +
theme(axis.text.x = element_text(angle=45, hjust = 1))+
labs(x = "Area", y = "# of Bike Points by Area",
title = "Areas that have the Lowest # of Bike Points")
Conclusion
Areas where bike points are less frequent are a good target for new bike points as many of these locations have quite large populations. For example Stratford has a population of 36,666 yet only has 1 bike point.
Most economically deprived areas
I read data from https://www.trustforlondon.org.uk/data/poverty-borough/ detailing the poverty rates by London Borough After Housing Costs (AHC).
ldn_poverty <- read_csv("C:/Users/keren/Downloads/Poverty_ Ldn_bor_(2015_16-2019_20).csv")## Rows: 32 Columns: 6
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (4): London borough, Poverty rate (AHC), Poverty rate (AHC) CI Lower, Po...
## dbl (2): Number in poverty (AHC), Confidence Interval (+/-)
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.ldn_poverty <- as.data.frame(ldn_poverty)
head(ldn_poverty)## London borough Number in poverty (AHC) Confidence Interval (+/-)
## 1 Barking and Dagenham 410000 5e+04
## 2 Barnet 500000 7e+04
## 3 Bexley 290000 5e+04
## 4 Brent 660000 1e+05
## 5 Bromley 320000 5e+04
## 6 Camden 320000 7e+04
## Poverty rate (AHC) Poverty rate (AHC) CI Lower Poverty rate (AHC) CI Upper
## 1 29% 24% 34.10%
## 2 25% 21.40% 29.50%
## 3 29% 23.10% 36.20%
## 4 36% 28.20% 43.70%
## 5 17% 14% 20.20%
## 6 34% 25% 45.30%summary(ldn_poverty)## London borough Number in poverty (AHC) Confidence Interval (+/-)
## Length:32 Min. :120000 Min. : 30000
## Class :character 1st Qu.:285000 1st Qu.: 50000
## Mode :character Median :380000 Median : 60000
## Mean :384194 Mean : 65806
## 3rd Qu.:490000 3rd Qu.: 80000
## Max. :690000 Max. :140000
## NA's :1 NA's :1
## Poverty rate (AHC) Poverty rate (AHC) CI Lower Poverty rate (AHC) CI Upper
## Length:32 Length:32 Length:32
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
##
##
##
## #renaming the columns
names(ldn_poverty) <- c('LondonBorough', 'No_in_Poverty_AHC', 'CI',
'PovertyRate_AHC', 'PovertyRateAHC_LowerCI', 'PovertyRateAHC_UpperCI')
head(ldn_poverty)## LondonBorough No_in_Poverty_AHC CI PovertyRate_AHC
## 1 Barking and Dagenham 410000 5e+04 29%
## 2 Barnet 500000 7e+04 25%
## 3 Bexley 290000 5e+04 29%
## 4 Brent 660000 1e+05 36%
## 5 Bromley 320000 5e+04 17%
## 6 Camden 320000 7e+04 34%
## PovertyRateAHC_LowerCI PovertyRateAHC_UpperCI
## 1 24% 34.10%
## 2 21.40% 29.50%
## 3 23.10% 36.20%
## 4 28.20% 43.70%
## 5 14% 20.20%
## 6 25% 45.30%#order by poverty rates
ldn_poverty_ordered <- ldn_poverty[order(ldn_poverty$PovertyRate_AHC, decreasing = TRUE),]#Boroughs with the highest poverty rate
top_10_poverty <- ldn_poverty_ordered[1:10,]
top_10_poverty## LondonBorough No_in_Poverty_AHC CI PovertyRate_AHC
## 29 Tower Hamlets 600000 80000 39%
## 4 Brent 660000 100000 36%
## 24 Newham 490000 80000 36%
## 32 Westminster 250000 50000 36%
## 13 Haringey 660000 110000 35%
## 22 Lewisham 500000 90000 35%
## 6 Camden 320000 70000 34%
## 9 Enfield 690000 80000 34%
## 31 Wandsworth 460000 140000 34%
## 8 Ealing 300000 50000 30%
## PovertyRateAHC_LowerCI PovertyRateAHC_UpperCI
## 29 32.50% 45.90%
## 4 28.20% 43.70%
## 24 29.40% 44.50%
## 32 27% 46.90%
## 13 27.20% 42.80%
## 22 26.30% 44.40%
## 6 25% 45.30%
## 9 28.90% 39.50%
## 31 22.10% 49.20%
## 8 23.80% 36.80%top_10_poverty %>%
ggplot(aes(x = LondonBorough, y = PovertyRate_AHC)) +
geom_bar(stat = 'identity', fill = 'blue') +
theme(axis.text.x = element_text(angle=45, hjust = 1))+
labs(x = "Borough (London)", y = "Poverty Rate (AHC)",
title = "London Boroughs with the Highest Poverty Rate (AHC)")
Conclusion
Economically deprived areas could benefit significantly from new bike points as lower income families may not be able to afford to use gym services or have the housing space to hold a bike.
Obesity related hospital admissions by London borough
Below I read data from https://digital.nhs.uk/data-and-information/publications/statistical/statistics-on-obesity-physical-activity-and-diet/england-2021/data-tables detailing the numbe of obesity related hospital admissions by borough.
Obesity_London <- read_csv("C:/Users/keren/Downloads/Obesity_rates_London.csv")## New names:
## Rows: 35 Columns: 7
## ── Column specification
## ──────────────────────────────────────────────────────── Delimiter: "," chr
## (1): ...1 dbl (6): No. of admissions: All persons, No. of admissions: Male, No.
## of adm...
## ℹ Use `spec()` to retrieve the full column specification for this data. ℹ
## Specify the column types or set `show_col_types = FALSE` to quiet this message.
## • `` -> `...1`Obesity_London <- as.data.frame(Obesity_London)
head(Obesity_London)## ...1 No. of admissions: All persons No. of admissions: Male
## 1 Barking and Dagenham 50 10
## 2 Barnet 80 35
## 3 Bexley 105 20
## 4 Brent 105 15
## 5 Bromley 85 20
## 6 Camden 50 15
## No. of admissions: Female Admissions per 100,000: All persons
## 1 40 24
## 2 50 21
## 3 85 43
## 4 85 34
## 5 65 27
## 6 35 21
## Admissions per 100,000:Male Admissions per 100,000:Female
## 1 10 38
## 2 18 25
## 3 18 67
## 4 11 57
## 5 14 39
## 6 10 32#renaming the columns
names(Obesity_London) <- c('LondonBorough', 'NOA_All_Persons', 'NOA_Male',
'NOA_All_Females', 'A_100K_All_Persons', 'A_100K_Male', 'A_100K_Female')
head(Obesity_London)## LondonBorough NOA_All_Persons NOA_Male NOA_All_Females
## 1 Barking and Dagenham 50 10 40
## 2 Barnet 80 35 50
## 3 Bexley 105 20 85
## 4 Brent 105 15 85
## 5 Bromley 85 20 65
## 6 Camden 50 15 35
## A_100K_All_Persons A_100K_Male A_100K_Female
## 1 24 10 38
## 2 21 18 25
## 3 43 18 67
## 4 34 11 57
## 5 27 14 39
## 6 21 10 32#order by obesity related admissions per 100,000 persons
Obesity_London_ordered <- Obesity_London[order(Obesity_London$A_100K_All_Persons, decreasing = TRUE),]
head(Obesity_London_ordered)## LondonBorough NOA_All_Persons NOA_Male NOA_All_Females A_100K_All_Persons
## 28 Southwark 200 40 160 70
## 23 Lewisham 185 35 150 62
## 22 Lambeth 145 30 120 49
## 3 Bexley 105 20 85 43
## 11 Greenwich 105 25 80 41
## 9 Ealing 125 35 90 39
## A_100K_Male A_100K_Female
## 28 30 109
## 23 28 95
## 22 21 79
## 3 18 67
## 11 21 61
## 9 22 56#Boroughs with most obesity related hospital admissions
top_10_obesity_admissions <- Obesity_London_ordered[1:10,]
top_10_obesity_admissions## LondonBorough NOA_All_Persons NOA_Male NOA_All_Females A_100K_All_Persons
## 28 Southwark 200 40 160 70
## 23 Lewisham 185 35 150 62
## 22 Lambeth 145 30 120 49
## 3 Bexley 105 20 85 43
## 11 Greenwich 105 25 80 41
## 9 Ealing 125 35 90 39
## 17 Hillingdon 105 35 70 38
## 18 Hounslow 90 25 60 37
## 4 Brent 105 15 85 34
## 8 Croydon 110 30 80 29
## A_100K_Male A_100K_Female
## 28 30 109
## 23 28 95
## 22 21 79
## 3 18 67
## 11 21 61
## 9 22 56
## 17 27 50
## 18 23 50
## 4 11 57
## 8 16 42top_10_obesity_admissions %>%
ggplot(aes(x = LondonBorough, y = A_100K_All_Persons))+
geom_bar(stat = 'identity', fill = 'purple')+
theme(axis.text.x = element_text(angle=45, hjust =1))+
labs(x = 'Borough (London)', y = 'Obesity Related Hospital Admissions per 100,000',
title = 'London Boroughs with the Rate of Obesity Related Hospital Admissions')
Conclusion
Boroughs with the highest rate of obesity related hospital admissions could stand to benefit tremendously from new bike points as exercising is one of the key ways to remain healthy.
Final thoughts
I believe that the areas highlighted throughout my report would be the key areas to look at when building new bike points as these areas could stand to benefit the most from increased availablity of bike points.