00-Human-WellBeing-Outcomes-Analysis
Pooja Choksi
2024-10-21
These chunks of code are used to produce all the figures in the main and supplementary manuscript.
#LOAD LIBRARIES
library(ggplot2)
library(reshape2)
library(dplyr)
library(tidyr)
library(tidyverse)
library(stringr)
library(highcharter)
library(networkD3)
library(wesanderson)
library(circlize)
options(repos = list(CRAN="http://cran.rstudio.com/"))
install.packages("knitr")##
## There are binary versions available but the source versions are later:
## binary source needs_compilation
## xfun 0.39 0.50 TRUE
## knitr 1.42 1.49 FALSE
##
## Binaries will be installed
## package 'xfun' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\pooja\AppData\Local\Temp\RtmpqS07Ok\downloaded_packageslibrary(knitr)
install.packages("tinytex")##
## There are binary versions available but the source versions are later:
## binary source needs_compilation
## xfun 0.39 0.50 TRUE
## tinytex 0.45 0.54 FALSE
##
## Binaries will be installed
## package 'xfun' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\pooja\AppData\Local\Temp\RtmpqS07Ok\downloaded_packagestinytex::install_tinytex()#LOAD DATA
#Load Final dataset coded by team
coded_data = read.csv("ToF-Review-Checked-Code-V3.csv")
#Load Baell's list of predatory journals downloaded from the online list
baells = read.csv("ToF-Review-Baells-List.csv")#FIGURE 3B: YEAR WISE PUBLICATIONS IN EACH COUNTRY This chunk of code produces Figure 3B in the manuscript which shows the number of publications from each of the countries over the years 1980 to mid-2023.
#function to split the outcomes data into separate columns (of 0 and 1, where 1 is if the outcome was experienced and if not -0)
split_and_encode <- function(data, column_name) {
data %>%
separate_rows(column_name, sep = ";") %>%
mutate(value = 1) %>%
pivot_wider(names_from = column_name, values_from = value, values_fill = list(value = 0))
}
#NUMBER OF STUDIES IN EACH COUNTRY
#one study was carried out in multiple countries, so first, we split this column
coded_data_split_countries <- split_and_encode(coded_data, "X3_1")
#the above code makes a 0-1 matrix for each country
#we select only the necessary columns for each country only
country_summary = select(coded_data_split_countries, c("X1_16_a_UID", "X1_4_Published_Year", "India", "Bangladesh", "Sri Lanka", "Pakistan", "Nepal"))
#pivot this above table from wide to long
country_summary_long = country_summary%>%tidyr::pivot_longer(!c(X1_16_a_UID, "X1_4_Published_Year"), names_to = "Country", values_to = c("Presence"))
#remove the zeros from the table
country_summary_long_ = filter(country_summary_long, Presence==1)
#aggregate the country by year
country_aggregated = country_summary_long_%>%group_by(Country, X1_4_Published_Year)%>%count()
#aggregate the data to get the total number of publications per year (across all countries)
country_aggregated_total <- country_aggregated %>%
group_by(X1_4_Published_Year) %>%
summarise(total_n = sum(n))
library(wesanderson)
# Plot of publications in every year in each South Asian country
country_year_plot <- ggplot(country_aggregated, aes(x = X1_4_Published_Year, y = n, fill = Country)) +
geom_bar(stat = "identity", position = "stack") + # Stacked bars using the 'n' value
xlab("Year") +
ylab("Number of Publications") +
labs(fill = "Country") +
theme_classic() +
theme(legend.position = "bottom") +
scale_fill_manual(name = "Country", values = wes_palette("Darjeeling1", n = 5)) +
scale_x_continuous(breaks = seq(1983, 2023, by = 5)) +
scale_y_continuous(breaks = seq(0, max(country_aggregated_total$total_n), by = 3),
limits = c(0, max(country_aggregated_total$total_n) * 1.1)) +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
#Plot
country_year_plot
#FIGURE 4: WHICH TOF SYSTEMS EXIST in SOUTH ASIA AND WHAT IS THEIR FREQEUNCY? This chunk of code produces Figure 4 in the manuscript. The colors represent the country in which the system was studied.
#function to split the outcomes data into separate columns (of 0 and 1, where 1 is if the outcome was experienced and if not -0)
split_and_encode <- function(data, column_name) {
data %>%
separate_rows(column_name, sep = ",") %>%
mutate(value = 1) %>%
pivot_wider(names_from = column_name, values_from = value, values_fill = list(value = 0))
}
#TOF SYSTEM STUDIED IN EACH COUNTRY
#split up the different kinds of ToF systems studied
coded_data_split_systems <- split_and_encode(coded_data, "X3_4")
#function to split the outcomes data into separate columns (of 0 and 1, where 1 is if the outcome was experienced and if not -0)
split_and_encode <- function(data, column_name) {
data %>%
separate_rows(column_name, sep = ";") %>%
mutate(value = 1) %>%
pivot_wider(names_from = column_name, values_from = value, values_fill = list(value = 0))
}
#one study was carried out in multiple countries, so first, we split this column
coded_data_split_systems_countries <- split_and_encode(coded_data_split_systems , "X3_1")
#the above code makes a 0-1 matrix for each system and for each country
#we select only the necessary columns for each system and country
systems_country_summary = select(coded_data_split_systems_countries, c("X1_16_a_UID", "India",
"Bangladesh", "Nepal",
"Pakistan", "Sri Lanka",
"Monoculture plantation", "Mixed plantation", "Multipurpose trees on farms or lots",
"shelterbelt/ hedgerow/ windbreaks", "tree/ forest (home)garden",
"Mixed plantation crops", "Alley cropping", "Orchard",
"Integrated production of animals (meat and dairy)/ crops and wood",
"Aqua-silvo-fishery", "Tree or shrub pasturelands", "Entomoforestry",
"Afforestation", "Reforestation", "Improved or fallow rotation",
"Meadow orchards", "Riparian buffer strip", "Wooded pasture products", "Unknown"))
#to pivot columns, first specify which are countries and which are systems
country_columns <- c("India","Bangladesh", "Nepal", "Pakistan", "Sri Lanka")
tof_columns <- c("Monoculture plantation", "Mixed plantation", "Multipurpose trees on farms or lots",
"shelterbelt/ hedgerow/ windbreaks", "tree/ forest (home)garden",
"Mixed plantation crops", "Alley cropping", "Orchard",
"Integrated production of animals (meat and dairy)/ crops and wood",
"Aqua-silvo-fishery", "Tree or shrub pasturelands", "Entomoforestry",
"Afforestation", "Reforestation", "Improved or fallow rotation",
"Meadow orchards", "Riparian buffer strip", "Wooded pasture products", "Unknown")
# Pivot longer for countries
long_country <- systems_country_summary %>%
pivot_longer(cols = all_of(country_columns), names_to = "Country", values_to = "Presence_Country") %>%
filter(Presence_Country == 1) # Keep only rows where country is present
#next, pivot longer for tof systems
long_tof_systems <- systems_country_summary %>%
pivot_longer(cols = all_of(tof_columns), names_to = "TOF_Type", values_to = "Presence_TOF_Type") %>%
filter(Presence_TOF_Type == 1) # Keep only rows where forest type is present
#combine these two long tables by UID
df_combined <- full_join(long_country, long_tof_systems, by = "X1_16_a_UID")
#aggregate TOF system by country
df_aggrgated = df_combined%>%group_by(Country, TOF_Type)%>%count()
#to plot recode the ToF system column
df_aggrgated <- df_aggrgated %>%
mutate(TOF_Type = recode(TOF_Type,
"Multipurpose trees on farms or lots" = "Multipurpose trees on farms",
"shelterbelt/ hedgerow/ windbreaks" = "Shelterbelts",
"tree/ forest (home)garden" = "Tree/Forest garden",
"Integrated production of animals (meat and dairy)/ crops and wood" = "Integrated production"))
#arrange in descending order
df_aggrgated_ordered <- df_aggrgated %>%
group_by(TOF_Type) %>%
summarise(total_count = sum(n)) %>%
arrange(desc(total_count)) %>%
mutate(TOF_Type = factor(TOF_Type, levels = TOF_Type)) %>%
left_join(df_aggrgated, by = "TOF_Type")
#recode everything under 10 observations
#see TOF_systems_aggregated_ordered. Recode entomoforestry, meadow orchards, unknown, wooded pasture products, Riparian buffer strip, Aqua-silvo-fishery
df_aggrgated_ordered <- df_aggrgated_ordered %>%
mutate(TOF_Type = recode(TOF_Type,
"Unknown" = "Other", "Entomoforestry" = "Other", "Wooded pasture products" = "Other", "Riparian buffer strip" = "Other", "Aqua-silvo-fishery" = "Other", "Meadow orchards" = "Other"))
#Plot bars with countries as colours
tof_aggregated_plot = ggplot(df_aggrgated_ordered, aes(x= reorder(TOF_Type, -total_count), y= n, fill = Country)) + geom_col() + theme_classic()+ labs(
x = "Trees outside Forest System",
y = "Frequency") +
scale_fill_manual(name = "Country", values = wes_palette("Darjeeling1", n = 5))+
scale_x_discrete(labels = function(x) str_wrap(x, width = 14)) +
scale_y_continuous(breaks = seq(0, max(df_aggrgated_ordered$total_count), by = 10)) +# Wrap text with a width of 10 characters
theme_classic() +
theme(
legend.position = "bottom",
legend.text = element_text(size = 12), # Adjust legend text size
legend.title = element_text(size = 14), # Adjust legend title size
axis.title.x = element_text(size = 14), # Adjust x-axis title size
axis.title.y = element_text(size = 14), # Adjust y-axis title size
axis.text.x = element_text(angle = 45, hjust = 1, size =14),
axis.text.y = element_text(hjust = 1, size =14)
)
tof_aggregated_plot
#FIGURE 5A: SANKEY PLOT OF OUTCOMES AND THE SYSTEMS THEY ARE STUDIED IN This chunk of code produces Figure 5B in the manuscript. The sankey plot shows the number of publications that examined the seven human wellbeing outcomes in the different trees outside forest systems.
#First split the outcomes
#function to split the outcomes data into separate columns (of 0 and 1, where 1 is if the outcome was experienced and if not -0)
split_and_encode <- function(data, column_name) {
data %>%
separate_rows(column_name, sep = ",") %>%
mutate(value = 1) %>%
pivot_wider(names_from = column_name, values_from = value, values_fill = list(value = 0))
}
#WELLBEING OUTCOME STUDIED
#use the split code above to split all the buckets of outcomes we studied
coded_data_split_outcomes = split_and_encode(coded_data, "X4_1")
#the above code makes a 0-1 matrix for each human wellbeing outcome
#we select only the necessary columns for each outcome
outcomes_summary = select(coded_data_split_outcomes, c("X1_16_a_UID","X3_1", "Material and living standards related outcomes","Economic security related outcomes", "Social relationships related outcomes", "Education related outcomes", "Work and Leisure related outcomes","Agency and Political voice related outcomes", "Health related outcomes"))
#pivot this above table from wide to long
outcomes_summary_long = outcomes_summary%>%tidyr::pivot_longer(!c(X1_16_a_UID, X3_1), names_to = "Wellbeing_Outcome", values_to = c("Presence"))
#remove the zeros from the table
outcomes_summary_long = filter(outcomes_summary_long, Presence==1)
#TOF SYSTEM STUDIED IN EACH COUNTRY
#split up the different kinds of ToF systems studied
coded_data_split_systems <- split_and_encode(coded_data, "X3_4")
#the above code makes a 0-1 matrix for each system
#we select only the necessary columns for each system
systems_summary = select(coded_data_split_systems, c("X1_16_a_UID", "X3_1", "Monoculture plantation", "Mixed plantation", "Multipurpose trees on farms or lots", "shelterbelt/ hedgerow/ windbreaks", "tree/ forest (home)garden", "Mixed plantation crops", "Alley cropping", "Orchard", "Integrated production of animals (meat and dairy)/ crops and wood", "Aqua-silvo-fishery", "Tree or shrub pasturelands", "Entomoforestry", "Afforestation", "Reforestation", "Improved or fallow rotation", "Meadow orchards", "Riparian buffer strip", "Wooded pasture products", "Unknown"))
#pivot this above table from wide to long
systems_summary_long = systems_summary%>%tidyr::pivot_longer(!c(X1_16_a_UID, X3_1), names_to = "ToF_System", values_to = c("Presence"))
#Combine the systems and outcomes tables made above
systems_outcomes = full_join(outcomes_summary_long, systems_summary_long)## Joining, by = c("X1_16_a_UID", "X3_1", "Presence")#summarise outcomes studied in each system in each country
systems_outcomes_summary = systems_outcomes%>%group_by(ToF_System, Wellbeing_Outcome, X3_1)%>%summarise(across(everything(), sum))## `summarise()` has grouped output by 'ToF_System', 'Wellbeing_Outcome'. You can
## override using the `.groups` argument.#remove the zeros from the table
systems_outcomes_summary = filter(systems_outcomes_summary, Presence>=1)
#Use this above file to make a sankey diagram
#systems_outcomes_sankey = select(systems_outcomes, -c(X1_16_a_UID, #Presence))
#select only three columns
for_sankey = dplyr::select(systems_outcomes_summary, ToF_System, Wellbeing_Outcome, Presence)
#recode the systems with fewer than 10 pubs
for_sankey = for_sankey%>%mutate(ToF_System = recode(ToF_System,
"Unknown" = "Other", "Entomoforestry" = "Other", "Wooded pasture products" = "Other", "Riparian buffer strip" = "Other", "Aqua-silvo-fishery" = "Other", "Meadow orchards" = "Other"))
for_sankey = for_sankey%>%group_by(ToF_System, Wellbeing_Outcome)%>%summarise(across(everything(), sum))## `summarise()` has grouped output by 'ToF_System'. You can override using the
## `.groups` argument.# Create a color palette of 21 color-blind friendly colors
color_blind_friendly_palette <- c(
"Afforestation" = "#E69F00", # Orange
"Agency and Political voice related outcomes" = "#56B4E9", # Sky Blue
"Economic security related outcomes" = "#009E73", # Teal
"Alley cropping" = "#F0E442", # Yellow
"Integrated production of animals (meat and dairy)/ crops and wood" = "#0072B2", # Blue
"Education related outcomes" = "#D55E00", # Vermilion
"Improved or fallow rotation" = "#CC79A7", # Pink
"Mixed plantation" = "#999999", # Grey
"Mixed plantation crops" = "#F2A900", # Gold
"Health related outcomes" = "#007A33", # Dark Green
"Material and living standards related outcomes" = "#A30000", # Dark Red
"Monoculture plantation" = "#FFB6C1", # Light Pink
"Orchard" = "#8C8C8C", # Medium Grey
"shelterbelt/ hedgerow/ windbreaks" ="#C2C2C2", # Light Grey
"Multipurpose trees on farms or lots" = "#FF5733", # Bright Red
"Reforestation" ="#FFC300", # Bright Yellow
"Tree or shrub pasturelands" = "#DAF7A6", # Light Green
"Social relationships related outcomes" = "#581845", # Dark Purple
"tree/ forest (home)garden" ="#900C3F", # Maroon
"Work and Leisure related outcomes" = "#FF5733", # Bright Orange
#"#28B463", # Bright Green
"Other" = "#4B0082", # Indigo
"NA" = "#C70039" # Cherry Red"
)
nodes <- for_sankey %>%
pivot_longer(cols = c(ToF_System, Wellbeing_Outcome), names_to = "outcome_type", values_to = "outcome") %>%
select(outcome) %>%
distinct() %>%
mutate(id = row_number() - 1) %>%
mutate(color = color_blind_friendly_palette[outcome])
links <- for_sankey %>%
left_join(nodes %>% rename(outcome1_id = id, ToF_System = outcome), by = "ToF_System") %>%
left_join(nodes %>% rename(outcome2_id = id, Wellbeing_Outcome = outcome, target_color = color), by = "Wellbeing_Outcome")%>%
select(source = outcome1_id, target = outcome2_id, Presence, color = target_color)## Adding missing grouping variables: `ToF_System`#Sankey
sankey <- sankeyNetwork(
Links = links,
Nodes = nodes,
Source = "source",
Target = "target",
Value = "Presence",
NodeID = "outcome",
units = "Units",
NodeGroup = "outcome",
fontSize = 14,
iterations = 0,
LinkGroup = "color") # Pass colors for links## Links is a tbl_df. Converting to a plain data frame.
## Nodes is a tbl_df. Converting to a plain data frame.sankey$x$nodes <-
sankey$x$nodes %>%
mutate(is_source_node = name %in% links$source)
#Remove the labels, which we will manually add in PPT for better legibility
htmlwidgets::onRender(
sankey,
'
function(el, x) {
// Hide node labels by setting their opacity to 0
d3.select(el).selectAll(".node text").style("opacity", 0);
}
'
)# Plot Sankey diagram
print(sankey)
sankey #FIGURE 5B: CIRCOS PLOT OF THE CO-OCCURENCE OF CATEGORIES OF WELLBEING OUTCOMES This chunk of code produces Figure 5B, which is a Circos plot showing the co-occurence of the human wellbeing outcomes in the reviewed literature.
#function to split the outcomes data into separate columns (of 0 and 1, where 1 is if the outcome was experienced and if not -0)
split_and_encode <- function(data, column_name) {
data %>%
separate_rows(column_name, sep = ",") %>%
mutate(value = 1) %>%
pivot_wider(names_from = column_name, values_from = value, values_fill = list(value = 0))
}
#use the split code above to split all the buckets of outcomes we studied
coded_data_split_outcomes = split_and_encode(coded_data, "X4_1")
#the above code makes a 0-1 matrix for each human wellbeing outcome
#we select only the necessary columns for each outcome
outcomes_summary = select(coded_data_split_outcomes, c("X1_16_a_UID","X3_1", "Material and living standards related outcomes","Economic security related outcomes", "Social relationships related outcomes", "Education related outcomes", "Work and Leisure related outcomes","Agency and Political voice related outcomes", "Health related outcomes"))
# Prepare data for Circos plot
# Create a matrix of presence/absence
presence_matrix <- as.matrix(outcomes_summary[-c(1, 2)]) # Exclude ID and Country
rownames(presence_matrix) <- outcomes_summary$X3_1 # Set country names as row names
# Convert presence/absence to a more suitable format for Circos
presence_matrix[presence_matrix == 0] <- NA # Replace 0s with NA
colors <- c(
"Agency and Political voice related outcomes" = "#1f77b4",
"Economic security related outcomes" = "#ff7f0e",
"Health related outcomes" = "#2ca02c",
"Material and living standards related outcomes" = "#d62728",
"Social relationships related outcomes" = "#9467bd",
"Work and Leisure related outcomes" = "#8c564b",
"Education related outcomes" = "#FF0000")
# Create a presence-absence matrix for the outcomes
outcomes_summary <- outcomes_summary[-c(1, 2)] # Exclude ID and Country
presence_matrix <- as.matrix(outcomes_summary)
# Calculate co-occurrences
co_occurrence <- t(presence_matrix) %*% presence_matrix
diag(co_occurrence) <- 0 # Set diagonal to 0 to ignore self-loops
# Set row and column names
rownames(co_occurrence) <- colnames(co_occurrence) <- colnames(outcomes_summary)
# Create the Circos plot
circos.clear() # Clear previous settings if any
chordDiagram(co_occurrence, transparency = 0.5, grid.col = colors,
annotationTrack = "grid", preAllocateTracks = 1)
# Remove labels while keeping the bands
circos.trackPlotRegion(track.index = 1, bg.border = NA) # Just to ensure the track is set without labels
#FIGURE 6: DIRECTION OF WELLBEING OUTCOMES FOR ENTIRE DATASET (ALL LAND TENURES) This chunk of code produces Figure 6, which is a bar plots showing the direction of change in the human wellbeing outcomes in the literature reviewed. We have combined agency and political voice outcomes and social relationship outcomes for the purpose of this figure.
#FOR ALL DATA COMBINED
#MATERIAL OUTCOMES
fuelwood_outcomes_sum = coded_data%>%group_by(X4_2_a)%>%count()
fuelwood_outcomes_sum$outcome = "Fuelwood"
fuelwood_outcomes_sum = fuelwood_outcomes_sum%>%rename(Direction = X4_2_a)
biomass_outcomes_sum = coded_data%>%group_by(X4_2_b)%>%count()
biomass_outcomes_sum$outcome = "Biomass"
biomass_outcomes_sum = biomass_outcomes_sum%>%rename(Direction = X4_2_b)
fiber_outcome_sum = coded_data%>%group_by(X4_2_c)%>%count()
fiber_outcome_sum$outcome = "Fiber"
fiber_outcome_sum = fiber_outcome_sum%>%rename(Direction = X4_2_c)
housingmat_outcome_sum =
coded_data%>%group_by(X4_2_d)%>%count()
housingmat_outcome_sum$outcome = "Housing materials"
housingmat_outcome_sum = housingmat_outcome_sum%>%rename(Direction = X4_2_d)
water_outcome_sum =
coded_data%>%group_by(X4_2_e)%>%count()
water_outcome_sum$outcome = "Water availability"
water_outcome_sum = water_outcome_sum%>%rename(Direction = X4_2_e)
assets_outcome_sum =
coded_data%>%group_by(X4_2_g)%>%count()
assets_outcome_sum$outcome = "Assets"
assets_outcome_sum = assets_outcome_sum%>%rename(Direction = X4_2_g)
shelter_outcome_sum =
coded_data%>%group_by(X4_2_h)%>%count()
shelter_outcome_sum$outcome = "Adequete shelter"
shelter_outcome_sum = shelter_outcome_sum%>%rename(Direction = X4_2_h)
material_outcomes = rbind(shelter_outcome_sum, assets_outcome_sum, water_outcome_sum, housingmat_outcome_sum, fiber_outcome_sum, biomass_outcomes_sum, fuelwood_outcomes_sum)
material_outcomes$Direction = as.factor(material_outcomes$Direction)
#recode factors with spelling mistakes
material_outcomes = material_outcomes%>%mutate(Direction = fct_recode(Direction, "Mixed - increased for some, decreased for some" = "Mixed - increased for some, decreased for others", "Neutral" = "Nuetral", "Decrease" = "decrease", "Unreported" = "unreported"))
#Remove the factor which indicates no code
to_remove = material_outcomes%>%filter(Direction == "")
to_remove_2 = material_outcomes%>%filter(Direction == "Unreported")
material_outcomes = anti_join(material_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")material_outcomes = anti_join(material_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")material_outcomes$Direction <- droplevels(material_outcomes$Direction)
material_outcomes <- material_outcomes%>%drop_na(Direction)
colour_palette = c('Increase'='#009E73','Decrease'='#D55E00','Mixed - increased for some, decreased for some'='#999999','Neutral'='#56B4E9')
material_plot = ggplot(material_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Material outcomes")+ ylab("Number of publications")+ theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#HEALTH OUTCOMES
disease_outcomes_sum = coded_data%>%group_by(X4_2_i)%>%count()
disease_outcomes_sum$outcome = "Disease"
disease_outcomes_sum = disease_outcomes_sum%>%rename(Direction = X4_2_i)
medicine_outcomes_sum = coded_data%>%group_by(X4_2_j)%>%count()
medicine_outcomes_sum$outcome = "Medical care"
medicine_outcomes_sum = medicine_outcomes_sum%>%rename(Direction = X4_2_j)
mental_outcomes_sum = coded_data%>%group_by(X4_2_k)%>%count()
mental_outcomes_sum$outcome = "Mental health"
mental_outcomes_sum = mental_outcomes_sum%>%rename(Direction = X4_2_k)
long_outcomes_sum = coded_data%>%group_by(X4_2_l)%>%count()
long_outcomes_sum$outcome = "Longevity"
long_outcomes_sum = long_outcomes_sum%>%rename(Direction = X4_2_l)
insurance_outcomes_sum = coded_data%>%group_by(X4_2_m)%>%count()
insurance_outcomes_sum$outcome = "Insurance"
insurance_outcomes_sum = insurance_outcomes_sum%>%rename(Direction = X4_2_m)
diet_outcomes_sum = coded_data%>%group_by(X4_2_n)%>%count()
diet_outcomes_sum$outcome = "Diet"
diet_outcomes_sum = diet_outcomes_sum%>%rename(Direction = X4_2_n)
health_outcomes = rbind(diet_outcomes_sum, mental_outcomes_sum, insurance_outcomes_sum, long_outcomes_sum, medicine_outcomes_sum, disease_outcomes_sum)
health_outcomes$Direction = as.factor(health_outcomes$Direction)
levels(health_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "neutral"
## [6] "Neutral"
## [7] "Unreported"#recode factors with spelling mistakes
health_outcomes = health_outcomes%>%mutate(Direction = fct_recode(Direction, "Neutral" = "neutral"))
#Remove the factor which indicates no code
to_remove = health_outcomes%>%filter(Direction == "")
to_remove_2 = health_outcomes%>%filter(Direction == "Unreported")
health_outcomes = anti_join(health_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")health_outcomes = anti_join(health_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")health_outcomes$Direction <- droplevels(health_outcomes$Direction)
health_outcomes <- health_outcomes%>%drop_na(Direction)
colour_palette = c('Increase'='#009E73','Decrease'='#D55E00','Mixed - increased for some, decreased for some'='#999999','Neutral'='#56B4E9')
health_plot = ggplot(health_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Health outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#EDUCATIONAL OUTCOMES
education_outcomes_sum = coded_data%>%group_by(X4_2_o)%>%count()
education_outcomes_sum$outcome = "Education"
education_outcomes_sum = education_outcomes_sum%>%rename(Direction = X4_2_o)
vocation_outcomes_sum = coded_data%>%group_by(X4_2_p)%>%count()
vocation_outcomes_sum$outcome = "Vocational training"
vocation_outcomes_sum = vocation_outcomes_sum%>%rename(Direction = X4_2_p)
education_outcomes = rbind(education_outcomes_sum, vocation_outcomes_sum)
education_outcomes$Direction = as.factor(education_outcomes$Direction)
levels(education_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Neutral"
## [6] "Unreported"#Remove the factor which indicates no code
to_remove = education_outcomes%>%filter(Direction == "")
to_remove_2 = education_outcomes%>%filter(Direction == "Unreported")
education_outcomes = anti_join(education_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")education_outcomes = anti_join(education_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")education_outcomes$Direction <- droplevels(education_outcomes$Direction)
education_outcomes = education_outcomes%>%drop_na(Direction)
education_plot = ggplot(education_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Educational outcomes")+ ylab("Number of publications")+ theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#LEISURE OUTCOMES
nature_outcomes_sum = coded_data%>%group_by(X4_2_q)%>%count()
nature_outcomes_sum$outcome = "Access to nature"
nature_outcomes_sum = nature_outcomes_sum%>%rename(Direction = X4_2_q)
qualtiy_outcomes_sum = coded_data%>%group_by(X4_2_r)%>%count()
qualtiy_outcomes_sum$outcome = "Quality of life"
qualtiy_outcomes_sum = qualtiy_outcomes_sum%>%rename(Direction = X4_2_r)
recreation_outcomes_sum = coded_data%>%group_by(X4_2_s)%>%count()
recreation_outcomes_sum$outcome = "Recreation"
recreation_outcomes_sum = recreation_outcomes_sum%>%rename(Direction = X4_2_s)
belief_outcomes_sum = coded_data%>%group_by(X4_2_t)%>%count()
belief_outcomes_sum$outcome = "Belief system"
belief_outcomes_sum = belief_outcomes_sum%>%rename(Direction = X4_2_t)
move_outcomes_sum = coded_data%>%group_by(X4_2_u)%>%count()
move_outcomes_sum$outcome = "Freedom to move"
move_outcomes_sum = move_outcomes_sum%>%rename(Direction = X4_2_u)
leisure_outcomes = rbind(move_outcomes_sum, nature_outcomes_sum, belief_outcomes_sum, recreation_outcomes_sum, qualtiy_outcomes_sum)
leisure_outcomes$Direction = as.factor(leisure_outcomes$Direction)
levels(leisure_outcomes$Direction)## [1] ""
## [2] "Increase"
## [3] "Mixed - increased for some, decreased for some"
## [4] "Neutral"
## [5] "Unreported"#Remove the factor which indicates no code
to_remove = leisure_outcomes%>%filter(Direction == "")
to_remove_2 = leisure_outcomes%>%filter(Direction == "Unreported")
leisure_outcomes = anti_join(leisure_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")leisure_outcomes = anti_join(leisure_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")leisure_outcomes$Direction <- droplevels(leisure_outcomes$Direction)
leisure_outcomes = leisure_outcomes%>%drop_na(Direction)
leisure_plot = ggplot(leisure_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Leisure/Work outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))+
scale_y_continuous(breaks = seq(0, 10, by= 2))
#AGENCY OUTCOMES
political_outcomes_sum = coded_data%>%group_by(X4_2_v)%>%count()
political_outcomes_sum$outcome = "Political voice"
political_outcomes_sum = political_outcomes_sum%>%rename(Direction = X4_2_v)
recognition_outcomes_sum = coded_data%>%group_by(X4_2_w)%>%count()
recognition_outcomes_sum$outcome = "Local recognition"
recognition_outcomes_sum = recognition_outcomes_sum%>%rename(Direction = X4_2_w)
equity_outcomes_sum = coded_data%>%group_by(X4_2_x)%>%count()
equity_outcomes_sum$outcome = "Social equity"
equity_outcomes_sum = equity_outcomes_sum%>%rename(Direction = X4_2_x)
interactions_outcomes_sum = coded_data%>%group_by(X4_2_y)%>%count()
interactions_outcomes_sum$outcome = "Community interactions"
interactions_outcomes_sum = interactions_outcomes_sum%>%rename(Direction = X4_2_y)
trust_outcomes_sum = coded_data%>%group_by(X4_2_z)%>%count()
trust_outcomes_sum$outcome = "Trust"
trust_outcomes_sum = trust_outcomes_sum%>%rename(Direction = X4_2_z)
agency_outcomes = rbind(political_outcomes_sum, recognition_outcomes_sum, trust_outcomes_sum, interactions_outcomes_sum, equity_outcomes_sum)
agency_outcomes$Direction = as.factor(agency_outcomes$Direction)
levels(agency_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Neutral"
## [6] "Unreported"
## [7] "UNreported"agency_outcomes = agency_outcomes%>%mutate(Direction = fct_recode(Direction, "Unreported" = "UNreported"))
#Remove the factor which indicates no code
to_remove = agency_outcomes%>%filter(Direction == "")
to_remove_2 = agency_outcomes%>%filter(Direction == "Unreported")
agency_outcomes = anti_join(agency_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")agency_outcomes = anti_join(agency_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")agency_outcomes$Direction <- droplevels(agency_outcomes$Direction)
agency_outcomes = agency_outcomes%>%drop_na(Direction)
agency_plot = ggplot(agency_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Agency/ Social outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#ECONOMIC OUTCOMES
income_outcomes_sum = coded_data%>%group_by(X4_2_aa)%>%count()
income_outcomes_sum$outcome = "Employment/ income"
income_outcomes_sum = income_outcomes_sum%>%rename(Direction = X4_2_aa)
loan_outcomes_sum = coded_data%>%group_by(X4_2_ab)%>%count()
loan_outcomes_sum$outcome = "Access to loans"
loan_outcomes_sum = loan_outcomes_sum%>%rename(Direction = X4_2_ab)
debt_outcomes_sum = coded_data%>%group_by(X4_2_ac)%>%count()
debt_outcomes_sum$outcome = "Lack of debt"
debt_outcomes_sum = debt_outcomes_sum%>%rename(Direction = X4_2_ac)
savings_outcomes_sum = coded_data%>%group_by(X4_2_ad)%>%count()
savings_outcomes_sum$outcome = "Savings/ provision for dependents"
savings_outcomes_sum = savings_outcomes_sum%>%rename(Direction = X4_2_ad)
economic_outcomes = rbind(savings_outcomes_sum, debt_outcomes_sum, loan_outcomes_sum, income_outcomes_sum)
economic_outcomes$Direction = as.factor(economic_outcomes$Direction)
levels(economic_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Neutral"
## [6] "Unreported"#Remove the factor which indicates no code
to_remove = economic_outcomes%>%filter(Direction == "")
to_remove_2 = economic_outcomes%>%filter(Direction == "Unreported")
economic_outcomes = anti_join(economic_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")economic_outcomes = anti_join(economic_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")economic_outcomes$Direction <- droplevels(economic_outcomes$Direction)
economic_outcomes = economic_outcomes%>%drop_na(Direction)
economic_plot = ggplot(economic_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Economic outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#PLot all together
library(ggpubr)## Warning: package 'ggpubr' was built under R version 4.1.3benefits_plot = ggarrange(material_plot, leisure_plot, health_plot, agency_plot, economic_plot, education_plot, ncol = 2, nrow = 3, common.legend = TRUE)
benefits_plot
#FIGURE S1: DIRECTION OF WELLBEING OUTCOMES FOR COMMUNITY-LAND TENURE ONLY This chunk of code produces Figure S1, which is a bar plots showing the direction of change in the human wellbeing outcomes for publications focused on communally held land tenure alone. We have combined agency and political voice outcomes and social relationship outcomes in this figure.
#FOR COMMUNALLY HELD LAND
community_data = coded_data%>%dplyr::filter(X3_8 == "Community-held")
#MATERIAL OUTCOMES
fuelwood_outcomes_sum = community_data%>%group_by(X4_2_a)%>%count()
fuelwood_outcomes_sum$outcome = "Fuelwood"
fuelwood_outcomes_sum = fuelwood_outcomes_sum%>%rename(Direction = X4_2_a)
biomass_outcomes_sum = community_data%>%group_by(X4_2_b)%>%count()
biomass_outcomes_sum$outcome = "Biomass"
biomass_outcomes_sum = biomass_outcomes_sum%>%rename(Direction = X4_2_b)
fiber_outcome_sum = community_data%>%group_by(X4_2_c)%>%count()
fiber_outcome_sum$outcome = "Fiber"
fiber_outcome_sum = fiber_outcome_sum%>%rename(Direction = X4_2_c)
housingmat_outcome_sum =
community_data%>%group_by(X4_2_d)%>%count()
housingmat_outcome_sum$outcome = "Housing materials"
housingmat_outcome_sum = housingmat_outcome_sum%>%rename(Direction = X4_2_d)
water_outcome_sum =
community_data%>%group_by(X4_2_e)%>%count()
water_outcome_sum$outcome = "Water availability"
water_outcome_sum = water_outcome_sum%>%rename(Direction = X4_2_e)
assets_outcome_sum =
community_data%>%group_by(X4_2_g)%>%count()
assets_outcome_sum$outcome = "Assets"
assets_outcome_sum = assets_outcome_sum%>%rename(Direction = X4_2_g)
shelter_outcome_sum =
community_data%>%group_by(X4_2_h)%>%count()
shelter_outcome_sum$outcome = "Adequete shelter"
shelter_outcome_sum = shelter_outcome_sum%>%rename(Direction = X4_2_h)
material_outcomes = rbind(shelter_outcome_sum, assets_outcome_sum, water_outcome_sum, housingmat_outcome_sum, fiber_outcome_sum, biomass_outcomes_sum, fuelwood_outcomes_sum)
material_outcomes$Direction = as.factor(material_outcomes$Direction)
#recode factors with spelling mistakes
material_outcomes = material_outcomes%>%mutate(Direction = fct_recode(Direction, "Mixed - increased for some, decreased for some" = "Mixed - increased for some, decreased for others", "Neutral" = "Nuetral", "Decrease" = "decrease", "Unreported" = "unreported"))## Warning: Unknown levels in `f`: Nuetral, decrease, unreported
## Warning: Unknown levels in `f`: Nuetral, decrease, unreported
## Warning: Unknown levels in `f`: Nuetral, decrease, unreported
## Warning: Unknown levels in `f`: Nuetral, decrease, unreported
## Warning: Unknown levels in `f`: Nuetral, decrease, unreported
## Warning: Unknown levels in `f`: Nuetral, decrease, unreported
## Warning: Unknown levels in `f`: Nuetral, decrease, unreported#Remove the factor which indicates no code
to_remove = material_outcomes%>%filter(Direction == "")
to_remove_2 = material_outcomes%>%filter(Direction == "Unreported")
material_outcomes = anti_join(material_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")material_outcomes = anti_join(material_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")material_outcomes$Direction <- droplevels(material_outcomes$Direction)
material_outcomes <- material_outcomes%>%drop_na(Direction)
colour_palette = c('Increase'='#009E73','Decrease'='#D55E00','Mixed - increased for some, decreased for some'='#999999','Neutral'='#56B4E9')
material_plot = ggplot(material_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Material outcomes")+ ylab("Number of publications")+ theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#HEALTH OUTCOMES
disease_outcomes_sum = community_data%>%group_by(X4_2_i)%>%count()
disease_outcomes_sum$outcome = "Disease"
disease_outcomes_sum = disease_outcomes_sum%>%rename(Direction = X4_2_i)
medicine_outcomes_sum = community_data%>%group_by(X4_2_j)%>%count()
medicine_outcomes_sum$outcome = "Medical care"
medicine_outcomes_sum = medicine_outcomes_sum%>%rename(Direction = X4_2_j)
mental_outcomes_sum = community_data%>%group_by(X4_2_k)%>%count()
mental_outcomes_sum$outcome = "Mental health"
mental_outcomes_sum = mental_outcomes_sum%>%rename(Direction = X4_2_k)
long_outcomes_sum = community_data%>%group_by(X4_2_l)%>%count()
long_outcomes_sum$outcome = "Longevity"
long_outcomes_sum = long_outcomes_sum%>%rename(Direction = X4_2_l)
insurance_outcomes_sum = community_data%>%group_by(X4_2_m)%>%count()
insurance_outcomes_sum$outcome = "Insurance"
insurance_outcomes_sum = insurance_outcomes_sum%>%rename(Direction = X4_2_m)
diet_outcomes_sum = community_data%>%group_by(X4_2_n)%>%count()
diet_outcomes_sum$outcome = "Diet"
diet_outcomes_sum = diet_outcomes_sum%>%rename(Direction = X4_2_n)
health_outcomes = rbind(diet_outcomes_sum, mental_outcomes_sum, insurance_outcomes_sum, long_outcomes_sum, medicine_outcomes_sum, disease_outcomes_sum)
health_outcomes$Direction = as.factor(health_outcomes$Direction)
levels(health_outcomes$Direction)## [1] "" "Increase" "neutral" "Unreported"#recode factors with spelling mistakes
health_outcomes = health_outcomes%>%mutate(Direction = fct_recode(Direction, "Neutral" = "neutral"))
#Remove the factor which indicates no code
to_remove = health_outcomes%>%filter(Direction == "")
to_remove_2 = health_outcomes%>%filter(Direction == "Unreported")
health_outcomes = anti_join(health_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")health_outcomes = anti_join(health_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")health_outcomes$Direction <- droplevels(health_outcomes$Direction)
health_outcomes <- health_outcomes%>%drop_na(Direction)
colour_palette = c('Increase'='#009E73','Decrease'='#D55E00','Mixed - increased for some, decreased for some'='#999999','Neutral'='#56B4E9')
health_plot = ggplot(health_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Health outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#EDUCATIONAL OUTCOMES
education_outcomes_sum = community_data%>%group_by(X4_2_o)%>%count()
education_outcomes_sum$outcome = "Education"
education_outcomes_sum = education_outcomes_sum%>%rename(Direction = X4_2_o)
vocation_outcomes_sum = community_data%>%group_by(X4_2_p)%>%count()
vocation_outcomes_sum$outcome = "Vocational training"
vocation_outcomes_sum = vocation_outcomes_sum%>%rename(Direction = X4_2_p)
education_outcomes = rbind(education_outcomes_sum, vocation_outcomes_sum)
education_outcomes$Direction = as.factor(education_outcomes$Direction)
levels(education_outcomes$Direction)## [1] "" "Increase" "Unreported"#Remove the factor which indicates no code
to_remove = education_outcomes%>%filter(Direction == "")
to_remove_2 = education_outcomes%>%filter(Direction == "Unreported")
education_outcomes = anti_join(education_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")education_outcomes = anti_join(education_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")education_outcomes$Direction <- droplevels(education_outcomes$Direction)
education_outcomes = education_outcomes%>%drop_na(Direction)
education_plot = ggplot(education_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Educational outcomes")+ ylab("Number of publications")+ theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#LEISURE OUTCOMES
nature_outcomes_sum = community_data%>%group_by(X4_2_q)%>%count()
nature_outcomes_sum$outcome = "Access to nature"
nature_outcomes_sum = nature_outcomes_sum%>%rename(Direction = X4_2_q)
qualtiy_outcomes_sum = community_data%>%group_by(X4_2_r)%>%count()
qualtiy_outcomes_sum$outcome = "Quality of life"
qualtiy_outcomes_sum = qualtiy_outcomes_sum%>%rename(Direction = X4_2_r)
recreation_outcomes_sum = community_data%>%group_by(X4_2_s)%>%count()
recreation_outcomes_sum$outcome = "Recreation"
recreation_outcomes_sum = recreation_outcomes_sum%>%rename(Direction = X4_2_s)
belief_outcomes_sum = community_data%>%group_by(X4_2_t)%>%count()
belief_outcomes_sum$outcome = "Belief system"
belief_outcomes_sum = belief_outcomes_sum%>%rename(Direction = X4_2_t)
move_outcomes_sum = community_data%>%group_by(X4_2_u)%>%count()
move_outcomes_sum$outcome = "Freedom to move"
move_outcomes_sum = move_outcomes_sum%>%rename(Direction = X4_2_u)
leisure_outcomes = rbind(move_outcomes_sum, nature_outcomes_sum, belief_outcomes_sum, recreation_outcomes_sum, qualtiy_outcomes_sum)
leisure_outcomes$Direction = as.factor(leisure_outcomes$Direction)
levels(leisure_outcomes$Direction)## [1] ""
## [2] "Increase"
## [3] "Mixed - increased for some, decreased for some"
## [4] "Unreported"#Remove the factor which indicates no code
to_remove = leisure_outcomes%>%filter(Direction == "")
to_remove_2 = leisure_outcomes%>%filter(Direction == "Unreported")
leisure_outcomes = anti_join(leisure_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")leisure_outcomes = anti_join(leisure_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")leisure_outcomes$Direction <- droplevels(leisure_outcomes$Direction)
leisure_outcomes = leisure_outcomes%>%drop_na(Direction)
leisure_plot = ggplot(leisure_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Leisure/Work outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))+
scale_y_continuous(breaks = seq(0, 10, by= 2))
#AGENCY OUTCOMES
political_outcomes_sum = community_data%>%group_by(X4_2_v)%>%count()
political_outcomes_sum$outcome = "Political voice"
political_outcomes_sum = political_outcomes_sum%>%rename(Direction = X4_2_v)
recognition_outcomes_sum = community_data%>%group_by(X4_2_w)%>%count()
recognition_outcomes_sum$outcome = "Local recognition"
recognition_outcomes_sum = recognition_outcomes_sum%>%rename(Direction = X4_2_w)
equity_outcomes_sum = community_data%>%group_by(X4_2_x)%>%count()
equity_outcomes_sum$outcome = "Social equity"
equity_outcomes_sum = equity_outcomes_sum%>%rename(Direction = X4_2_x)
interactions_outcomes_sum = community_data%>%group_by(X4_2_y)%>%count()
interactions_outcomes_sum$outcome = "Community interactions"
interactions_outcomes_sum = interactions_outcomes_sum%>%rename(Direction = X4_2_y)
trust_outcomes_sum = community_data%>%group_by(X4_2_z)%>%count()
trust_outcomes_sum$outcome = "Trust"
trust_outcomes_sum = trust_outcomes_sum%>%rename(Direction = X4_2_z)
agency_outcomes = rbind(political_outcomes_sum, recognition_outcomes_sum, trust_outcomes_sum, interactions_outcomes_sum, equity_outcomes_sum)
agency_outcomes$Direction = as.factor(agency_outcomes$Direction)
levels(agency_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Unreported"agency_outcomes = agency_outcomes%>%mutate(Direction = fct_recode(Direction, "Unreported" = "UNreported"))## Warning: Unknown levels in `f`: UNreported## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported#Remove the factor which indicates no code
to_remove = agency_outcomes%>%filter(Direction == "")
to_remove_2 = agency_outcomes%>%filter(Direction == "Unreported")
agency_outcomes = anti_join(agency_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")agency_outcomes = anti_join(agency_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")agency_outcomes$Direction <- droplevels(agency_outcomes$Direction)
agency_outcomes = agency_outcomes%>%drop_na(Direction)
agency_plot = ggplot(agency_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Agency/ Social outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#ECONOMIC OUTCOMES
income_outcomes_sum = community_data%>%group_by(X4_2_aa)%>%count()
income_outcomes_sum$outcome = "Employment/ income"
income_outcomes_sum = income_outcomes_sum%>%rename(Direction = X4_2_aa)
loan_outcomes_sum = community_data%>%group_by(X4_2_ab)%>%count()
loan_outcomes_sum$outcome = "Access to loans"
loan_outcomes_sum = loan_outcomes_sum%>%rename(Direction = X4_2_ab)
debt_outcomes_sum = community_data%>%group_by(X4_2_ac)%>%count()
debt_outcomes_sum$outcome = "Lack of debt"
debt_outcomes_sum = debt_outcomes_sum%>%rename(Direction = X4_2_ac)
savings_outcomes_sum = community_data%>%group_by(X4_2_ad)%>%count()
savings_outcomes_sum$outcome = "Savings/ provision for dependents"
savings_outcomes_sum = savings_outcomes_sum%>%rename(Direction = X4_2_ad)
economic_outcomes = rbind(savings_outcomes_sum, debt_outcomes_sum, loan_outcomes_sum, income_outcomes_sum)
economic_outcomes$Direction = as.factor(economic_outcomes$Direction)
levels(economic_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Unreported"#Remove the factor which indicates no code
to_remove = economic_outcomes%>%filter(Direction == "")
to_remove_2 = economic_outcomes%>%filter(Direction == "Unreported")
economic_outcomes = anti_join(economic_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")economic_outcomes = anti_join(economic_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")economic_outcomes$Direction <- droplevels(economic_outcomes$Direction)
economic_outcomes = economic_outcomes%>%drop_na(Direction)
economic_plot = ggplot(economic_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Economic outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#PLot all together
library(ggpubr)
community_benefits_plot = ggarrange(material_plot, leisure_plot, health_plot, agency_plot, economic_plot, education_plot, ncol = 2, nrow = 3, common.legend = TRUE)
community_benefits_plot
#FIGURE S2: DIRECTION OF WELLBEING OUTCOMES FOR PRIVATE LAND TENURE ONLY This chunk of code produces Figure S2, which is a bar plots showing the direction of change in the human wellbeing outcomes for publications focused on private land tenure alone. We have combined agency and political voice outcomes and social relationship outcomes in this figure.
#FOR PRIVATE LAND TENURE
private_data = coded_data%>%dplyr::filter(X3_8 == "Private")
#MATERIAL OUTCOMES
fuelwood_outcomes_sum = private_data%>%group_by(X4_2_a)%>%count()
fuelwood_outcomes_sum$outcome = "Fuelwood"
fuelwood_outcomes_sum = fuelwood_outcomes_sum%>%rename(Direction = X4_2_a)
biomass_outcomes_sum = private_data%>%group_by(X4_2_b)%>%count()
biomass_outcomes_sum$outcome = "Biomass"
biomass_outcomes_sum = biomass_outcomes_sum%>%rename(Direction = X4_2_b)
fiber_outcome_sum = private_data%>%group_by(X4_2_c)%>%count()
fiber_outcome_sum$outcome = "Fiber"
fiber_outcome_sum = fiber_outcome_sum%>%rename(Direction = X4_2_c)
housingmat_outcome_sum =
private_data%>%group_by(X4_2_d)%>%count()
housingmat_outcome_sum$outcome = "Housing materials"
housingmat_outcome_sum = housingmat_outcome_sum%>%rename(Direction = X4_2_d)
water_outcome_sum =
private_data%>%group_by(X4_2_e)%>%count()
water_outcome_sum$outcome = "Water availability"
water_outcome_sum = water_outcome_sum%>%rename(Direction = X4_2_e)
assets_outcome_sum =
private_data%>%group_by(X4_2_g)%>%count()
assets_outcome_sum$outcome = "Assets"
assets_outcome_sum = assets_outcome_sum%>%rename(Direction = X4_2_g)
shelter_outcome_sum =
private_data%>%group_by(X4_2_h)%>%count()
shelter_outcome_sum$outcome = "Adequete shelter"
shelter_outcome_sum = shelter_outcome_sum%>%rename(Direction = X4_2_h)
material_outcomes = rbind(shelter_outcome_sum, assets_outcome_sum, water_outcome_sum, housingmat_outcome_sum, fiber_outcome_sum, biomass_outcomes_sum, fuelwood_outcomes_sum)
material_outcomes$Direction = as.factor(material_outcomes$Direction)
#recode factors with spelling mistakes
material_outcomes = material_outcomes%>%mutate(Direction = fct_recode(Direction, "Mixed - increased for some, decreased for some" = "Mixed - increased for some, decreased for others", "Neutral" = "Nuetral", "Decrease" = "decrease", "Unreported" = "unreported"))## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease
## Warning: Unknown levels in `f`: decrease#Remove the factor which indicates no code
to_remove = material_outcomes%>%filter(Direction == "")
to_remove_2 = material_outcomes%>%filter(Direction == "Unreported")
material_outcomes = anti_join(material_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")material_outcomes = anti_join(material_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")material_outcomes$Direction <- droplevels(material_outcomes$Direction)
material_outcomes <- material_outcomes%>%drop_na(Direction)
colour_palette = c('Increase'='#009E73','Decrease'='#D55E00','Mixed - increased for some, decreased for some'='#999999','Neutral'='#56B4E9')
material_plot = ggplot(material_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Material outcomes")+ ylab("Number of publications")+ theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#HEALTH OUTCOMES
disease_outcomes_sum = private_data%>%group_by(X4_2_i)%>%count()
disease_outcomes_sum$outcome = "Disease"
disease_outcomes_sum = disease_outcomes_sum%>%rename(Direction = X4_2_i)
medicine_outcomes_sum = private_data%>%group_by(X4_2_j)%>%count()
medicine_outcomes_sum$outcome = "Medical care"
medicine_outcomes_sum = medicine_outcomes_sum%>%rename(Direction = X4_2_j)
mental_outcomes_sum = private_data%>%group_by(X4_2_k)%>%count()
mental_outcomes_sum$outcome = "Mental health"
mental_outcomes_sum = mental_outcomes_sum%>%rename(Direction = X4_2_k)
long_outcomes_sum = private_data%>%group_by(X4_2_l)%>%count()
long_outcomes_sum$outcome = "Longevity"
long_outcomes_sum = long_outcomes_sum%>%rename(Direction = X4_2_l)
insurance_outcomes_sum = private_data%>%group_by(X4_2_m)%>%count()
insurance_outcomes_sum$outcome = "Insurance"
insurance_outcomes_sum = insurance_outcomes_sum%>%rename(Direction = X4_2_m)
diet_outcomes_sum = private_data%>%group_by(X4_2_n)%>%count()
diet_outcomes_sum$outcome = "Diet"
diet_outcomes_sum = diet_outcomes_sum%>%rename(Direction = X4_2_n)
health_outcomes = rbind(diet_outcomes_sum, mental_outcomes_sum, insurance_outcomes_sum, long_outcomes_sum, medicine_outcomes_sum, disease_outcomes_sum)
health_outcomes$Direction = as.factor(health_outcomes$Direction)
levels(health_outcomes$Direction)## [1] ""
## [2] "Increase"
## [3] "Mixed - increased for some, decreased for some"
## [4] "neutral"
## [5] "Neutral"
## [6] "Unreported"#recode factors with spelling mistakes
health_outcomes = health_outcomes%>%mutate(Direction = fct_recode(Direction, "Neutral" = "neutral"))
#Remove the factor which indicates no code
to_remove = health_outcomes%>%filter(Direction == "")
to_remove_2 = health_outcomes%>%filter(Direction == "Unreported")
health_outcomes = anti_join(health_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")health_outcomes = anti_join(health_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")health_outcomes$Direction <- droplevels(health_outcomes$Direction)
health_outcomes <- health_outcomes%>%drop_na(Direction)
colour_palette = c('Increase'='#009E73','Decrease'='#D55E00','Mixed - increased for some, decreased for some'='#999999','Neutral'='#56B4E9')
health_plot = ggplot(health_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Health outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#EDUCATIONAL OUTCOMES
education_outcomes_sum = private_data%>%group_by(X4_2_o)%>%count()
education_outcomes_sum$outcome = "Education"
education_outcomes_sum = education_outcomes_sum%>%rename(Direction = X4_2_o)
vocation_outcomes_sum = private_data%>%group_by(X4_2_p)%>%count()
vocation_outcomes_sum$outcome = "Vocational training"
vocation_outcomes_sum = vocation_outcomes_sum%>%rename(Direction = X4_2_p)
education_outcomes = rbind(education_outcomes_sum, vocation_outcomes_sum)
education_outcomes$Direction = as.factor(education_outcomes$Direction)
levels(education_outcomes$Direction)## [1] ""
## [2] "Increase"
## [3] "Mixed - increased for some, decreased for some"
## [4] "Unreported"#Remove the factor which indicates no code
to_remove = education_outcomes%>%filter(Direction == "")
to_remove_2 = education_outcomes%>%filter(Direction == "Unreported")
education_outcomes = anti_join(education_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")education_outcomes = anti_join(education_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")education_outcomes$Direction <- droplevels(education_outcomes$Direction)
education_outcomes = education_outcomes%>%drop_na(Direction)
education_plot = ggplot(education_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Educational outcomes")+ ylab("Number of publications")+ theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#LEISURE OUTCOMES
nature_outcomes_sum = private_data%>%group_by(X4_2_q)%>%count()
nature_outcomes_sum$outcome = "Access to nature"
nature_outcomes_sum = nature_outcomes_sum%>%rename(Direction = X4_2_q)
qualtiy_outcomes_sum = private_data%>%group_by(X4_2_r)%>%count()
qualtiy_outcomes_sum$outcome = "Quality of life"
qualtiy_outcomes_sum = qualtiy_outcomes_sum%>%rename(Direction = X4_2_r)
recreation_outcomes_sum = private_data%>%group_by(X4_2_s)%>%count()
recreation_outcomes_sum$outcome = "Recreation"
recreation_outcomes_sum = recreation_outcomes_sum%>%rename(Direction = X4_2_s)
belief_outcomes_sum = private_data%>%group_by(X4_2_t)%>%count()
belief_outcomes_sum$outcome = "Belief system"
belief_outcomes_sum = belief_outcomes_sum%>%rename(Direction = X4_2_t)
move_outcomes_sum = private_data%>%group_by(X4_2_u)%>%count()
move_outcomes_sum$outcome = "Freedom to move"
move_outcomes_sum = move_outcomes_sum%>%rename(Direction = X4_2_u)
leisure_outcomes = rbind(move_outcomes_sum, nature_outcomes_sum, belief_outcomes_sum, recreation_outcomes_sum, qualtiy_outcomes_sum)
leisure_outcomes$Direction = as.factor(leisure_outcomes$Direction)
levels(leisure_outcomes$Direction)## [1] "" "Increase" "Neutral" "Unreported"#Remove the factor which indicates no code
to_remove = leisure_outcomes%>%filter(Direction == "")
to_remove_2 = leisure_outcomes%>%filter(Direction == "Unreported")
leisure_outcomes = anti_join(leisure_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")leisure_outcomes = anti_join(leisure_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")leisure_outcomes$Direction <- droplevels(leisure_outcomes$Direction)
leisure_outcomes = leisure_outcomes%>%drop_na(Direction)
leisure_plot = ggplot(leisure_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Leisure/Work outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))+
scale_y_continuous(breaks = seq(0, 10, by= 2))
#AGENCY OUTCOMES
political_outcomes_sum = private_data%>%group_by(X4_2_v)%>%count()
political_outcomes_sum$outcome = "Political voice"
political_outcomes_sum = political_outcomes_sum%>%rename(Direction = X4_2_v)
recognition_outcomes_sum = private_data%>%group_by(X4_2_w)%>%count()
recognition_outcomes_sum$outcome = "Local recognition"
recognition_outcomes_sum = recognition_outcomes_sum%>%rename(Direction = X4_2_w)
equity_outcomes_sum = private_data%>%group_by(X4_2_x)%>%count()
equity_outcomes_sum$outcome = "Social equity"
equity_outcomes_sum = equity_outcomes_sum%>%rename(Direction = X4_2_x)
interactions_outcomes_sum = private_data%>%group_by(X4_2_y)%>%count()
interactions_outcomes_sum$outcome = "Community interactions"
interactions_outcomes_sum = interactions_outcomes_sum%>%rename(Direction = X4_2_y)
trust_outcomes_sum = private_data%>%group_by(X4_2_z)%>%count()
trust_outcomes_sum$outcome = "Trust"
trust_outcomes_sum = trust_outcomes_sum%>%rename(Direction = X4_2_z)
agency_outcomes = rbind(political_outcomes_sum, recognition_outcomes_sum, trust_outcomes_sum, interactions_outcomes_sum, equity_outcomes_sum)
agency_outcomes$Direction = as.factor(agency_outcomes$Direction)
levels(agency_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Neutral"
## [6] "Unreported"agency_outcomes = agency_outcomes%>%mutate(Direction = fct_recode(Direction, "Unreported" = "UNreported"))## Warning: Unknown levels in `f`: UNreported## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported
## Warning: Unknown levels in `f`: UNreported#Remove the factor which indicates no code
to_remove = agency_outcomes%>%filter(Direction == "")
to_remove_2 = agency_outcomes%>%filter(Direction == "Unreported")
agency_outcomes = anti_join(agency_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")agency_outcomes = anti_join(agency_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")agency_outcomes$Direction <- droplevels(agency_outcomes$Direction)
agency_outcomes = agency_outcomes%>%drop_na(Direction)
agency_plot = ggplot(agency_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Agency/ Social outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#ECONOMIC OUTCOMES
income_outcomes_sum = private_data%>%group_by(X4_2_aa)%>%count()
income_outcomes_sum$outcome = "Employment/ income"
income_outcomes_sum = income_outcomes_sum%>%rename(Direction = X4_2_aa)
loan_outcomes_sum = private_data%>%group_by(X4_2_ab)%>%count()
loan_outcomes_sum$outcome = "Access to loans"
loan_outcomes_sum = loan_outcomes_sum%>%rename(Direction = X4_2_ab)
debt_outcomes_sum = private_data%>%group_by(X4_2_ac)%>%count()
debt_outcomes_sum$outcome = "Lack of debt"
debt_outcomes_sum = debt_outcomes_sum%>%rename(Direction = X4_2_ac)
savings_outcomes_sum = private_data%>%group_by(X4_2_ad)%>%count()
savings_outcomes_sum$outcome = "Savings/ provision for dependents"
savings_outcomes_sum = savings_outcomes_sum%>%rename(Direction = X4_2_ad)
economic_outcomes = rbind(savings_outcomes_sum, debt_outcomes_sum, loan_outcomes_sum, income_outcomes_sum)
economic_outcomes$Direction = as.factor(economic_outcomes$Direction)
levels(economic_outcomes$Direction)## [1] ""
## [2] "Decrease"
## [3] "Increase"
## [4] "Mixed - increased for some, decreased for some"
## [5] "Neutral"
## [6] "Unreported"#Remove the factor which indicates no code
to_remove = economic_outcomes%>%filter(Direction == "")
to_remove_2 = economic_outcomes%>%filter(Direction == "Unreported")
economic_outcomes = anti_join(economic_outcomes, to_remove)## Joining, by = c("Direction", "n", "outcome")economic_outcomes = anti_join(economic_outcomes, to_remove_2)## Joining, by = c("Direction", "n", "outcome")economic_outcomes$Direction <- droplevels(economic_outcomes$Direction)
economic_outcomes = economic_outcomes%>%drop_na(Direction)
economic_plot = ggplot(economic_outcomes, aes(x= factor(outcome), y= n, fill = factor(Direction)))+ geom_bar(position="stack", stat="identity") +xlab("Economic outcomes")+ ylab("Number of publications")+theme_classic()+ scale_fill_manual(name = "Direction", values = colour_palette) + coord_flip()+ scale_x_discrete(labels = function(x) str_wrap(x, width = 10))
#PLot all together
library(ggpubr)
private_benefits_plot = ggarrange(material_plot, leisure_plot, health_plot, agency_plot, economic_plot, education_plot, ncol = 2, nrow = 3, common.legend = TRUE)
private_benefits_plot
#FIGURE S3: CLOSER EXAMINATION OF AGENCY OUTCOMES AND TOF SYSTEMS THEY ARE EXPERIENCED IN This chunk of code produces S3, which is a bar plot of the change in direction of the Agency and social outcomes of trees outside forest systems linked to the ToF systems studied
#function to split the outcomes data into separate columns (of 0 and 1, where 1 is if the outcome was experienced and if not -0)
split_and_encode <- function(data, column_name) {
data %>%
separate_rows(column_name, sep = ",") %>%
mutate(value = 1) %>%
pivot_wider(names_from = column_name, values_from = value, values_fill = list(value = 0))
}
#TOF SYSTEM STUDIED
#split up the different kinds of ToF systems studied
coded_data_split_systems <- split_and_encode(coded_data, "X3_4")
#the above code makes a 0-1 matrix for each system
#we select only the necessary columns for each system
systems_summary = select(coded_data_split_systems, c("Monoculture plantation", "Mixed plantation", "Multipurpose trees on farms or lots", "shelterbelt/ hedgerow/ windbreaks", "tree/ forest (home)garden", "Mixed plantation crops", "Alley cropping", "Orchard", "Integrated production of animals (meat and dairy)/ crops and wood", "Aqua-silvo-fishery", "Tree or shrub pasturelands", "Entomoforestry", "Afforestation", "Reforestation", "Improved or fallow rotation", "Meadow orchards", "Riparian buffer strip", "Wooded pasture products", "Unknown", "X4_2_v", "X4_2_w", "X4_2_x", "X4_2_y", "X4_2_z"))
systems_summary_agency = systems_summary%>%group_by(`Monoculture plantation`, `Mixed plantation`, `Mixed plantation crops`, Afforestation, Reforestation, `Wooded pasture products`, `Multipurpose trees on farms or lots`, `tree/ forest (home)garden`, `Tree or shrub pasturelands`, `Alley cropping`, `shelterbelt/ hedgerow/ windbreaks`, `Integrated production of animals (meat and dairy)/ crops and wood`, Orchard, `Aqua-silvo-fishery`, Entomoforestry, `Improved or fallow rotation`, `Meadow orchards`, `Riparian buffer strip`, X4_2_v, X4_2_w, X4_2_x, X4_2_y, X4_2_z)%>%count()
# Pivot into long format
systems_summary_long <- systems_summary %>%
pivot_longer(
cols = c(`Monoculture plantation`, `Mixed plantation`, `Mixed plantation crops`, Afforestation, Reforestation, `Wooded pasture products`, `Multipurpose trees on farms or lots`, `tree/ forest (home)garden`, `Tree or shrub pasturelands`, `Alley cropping`, `shelterbelt/ hedgerow/ windbreaks`, `Integrated production of animals (meat and dairy)/ crops and wood`, Orchard, `Aqua-silvo-fishery`, Entomoforestry, `Improved or fallow rotation`, `Meadow orchards`, `Riparian buffer strip`,`Unknown`),
names_to = "TOF_System",
values_to = "Present")
#Pivot longer on the agency and social outcomes
systems_agency_outcomes_long = systems_summary_long %>%
tidyr::pivot_longer(cols = X4_2_v:X4_2_z,
names_to = "Agency_Social_outcomes",
values_to = "Direction",
values_drop_na = TRUE)%>%
filter(Present == 1) %>%
select(-Present) # Remove the 'present' column
#Rename the outcomes in the long table
systems_agency_outcomes_long <- systems_agency_outcomes_long %>%
mutate(Agency_Social_outcomes = case_when(
grepl("X4_2_v", Agency_Social_outcomes) ~ "Political_Voice",
grepl("X4_2_w", Agency_Social_outcomes) ~ "Local_Recognition",
grepl("X4_2_x", Agency_Social_outcomes) ~ "Social_Equity",
grepl("X4_2_y", Agency_Social_outcomes) ~ "Community_Interactions",
grepl("X4_2_z", Agency_Social_outcomes) ~ "Trust"
))
# Remove unwanted observations (Unreported and NAs)
to_remove = systems_agency_outcomes_long%>%filter(Direction == "")
to_remove_2 = systems_agency_outcomes_long%>%filter(Direction == "Unreported")
to_remove_3 = systems_agency_outcomes_long%>%filter(Direction == "UNreported")
systems_agency_outcomes_long = anti_join(systems_agency_outcomes_long, to_remove)## Joining, by = c("TOF_System", "Agency_Social_outcomes", "Direction")systems_agency_outcomes_long = anti_join(systems_agency_outcomes_long, to_remove_2)## Joining, by = c("TOF_System", "Agency_Social_outcomes", "Direction")systems_agency_outcomes_long = anti_join(systems_agency_outcomes_long, to_remove_3)## Joining, by = c("TOF_System", "Agency_Social_outcomes", "Direction")systems_agency_outcomes_long <- systems_agency_outcomes_long%>%drop_na(Direction)
#Relabel one of the labels - mixed
systems_agency_outcomes_long = systems_agency_outcomes_long%>%mutate(Direction = fct_recode(Direction, "Mixed" = "Mixed - increased for some, decreased for some"))
#SUm up the outcomes
#Filter each type of social outcomes
#1. Trust
Trust = systems_agency_outcomes_long%>%filter(Agency_Social_outcomes == "Trust")
Trust_sum = Trust%>%group_by(Direction, TOF_System)%>%count()
Trust_totals = Trust_sum%>%group_by(Direction)%>%summarise(total_n = sum(n), groups = 'drop')
# Join trsut total counts back to the Trust sums dataframe
Trust_percentage <- Trust_sum %>%
left_join(Trust_totals, by = "Direction") %>%
mutate(percentage = (n / total_n) * 100)
# Create a 12-color palette
install.packages("RColorBrewer")## Installing package into 'C:/Users/pooja/Documents/R/win-library/4.1'
## (as 'lib' is unspecified)## package 'RColorBrewer' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\pooja\AppData\Local\Temp\RtmpqS07Ok\downloaded_packageslibrary(RColorBrewer)## Warning: package 'RColorBrewer' was built under R version 4.1.3# Combine palettes to create a custom 18-color palette
# Get palettes with a maximum number of colors
palette1 <- brewer.pal(n = 9, name = "Set1") # 9 colors
palette2 <- brewer.pal(n = 9, name = "Set2") # 9 colors## Warning in brewer.pal(n = 9, name = "Set2"): n too large, allowed maximum for palette Set2 is 8
## Returning the palette you asked for with that many colors# Combine them into a single palette
color_palette_18 <- c(palette1, palette2)
# Create a stacked bar plot
Trust_plot = ggplot(Trust_percentage, aes(x = Direction, y = percentage, fill = TOF_System)) +
geom_bar(stat = "identity", position = "stack") +
#geom_text(aes(label = n),
# position = position_stack(vjust = 0.5),
# size = 3.5,
# color = "black",
# check_overlap = TRUE) +
labs(
title = "Trust",
x = "Direction of Change",
y = "Percentage",
fill = "ToF System"
)+scale_fill_manual(values = color_palette_18) +
theme_classic() +
theme(
#legend.position = "bottom",
axis.text.x = element_text(angle = 45, hjust = 1) # Rotate x-axis labels for better readability
)
#2. Social_Equity
Social_Equity = systems_agency_outcomes_long%>%filter(Agency_Social_outcomes == "Social_Equity")
Social_Equity_sum = Social_Equity%>%group_by(Direction, TOF_System)%>%count()
Social_Equity_totals = Social_Equity_sum%>%group_by(Direction)%>%summarise(total_n = sum(n), groups = 'drop')
# Join social equity total counts back to theSocial_Equity sums dataframe
Social_Equity_percentage <- Social_Equity_sum %>%
left_join(Social_Equity_totals, by = "Direction") %>%
mutate(percentage = (n / total_n) * 100)
# Create a stacked bar plot
Social_Equity_plot = ggplot(Social_Equity_percentage, aes(x = Direction, y = percentage, fill = TOF_System)) +
geom_bar(stat = "identity", position = "stack") +
#geom_text(aes(label = n),
# position = position_stack(vjust = 0.5),
# size = 3.5,
# color = "black",
# check_overlap = TRUE) +
labs(
title = "Social Equity",
x = "Direction of Change",
y = "Percentage",
fill = "ToF System"
)+scale_fill_manual(values = color_palette_18) + # Use the viridis palette
theme_classic() +
theme(
#legend.position = "bottom",
axis.text.x = element_text(angle = 45, hjust = 1) # Rotate x-axis labels for better readability
)
#3. political_voice
political_voice = systems_agency_outcomes_long%>%filter(Agency_Social_outcomes == "Political_Voice")
political_voice_sum = political_voice%>%group_by(Direction, TOF_System)%>%count()
political_voice_totals = political_voice_sum%>%group_by(Direction)%>%summarise(total_n = sum(n), groups = 'drop')
# Join social equity total counts back to thepolitical_voice sums dataframe
political_voice_percentage <- political_voice_sum %>%
left_join(political_voice_totals, by = "Direction") %>%
mutate(percentage = (n / total_n) * 100)
# Create a stacked bar plot
political_voice_plot = ggplot(political_voice_percentage, aes(x = Direction, y = percentage, fill = TOF_System)) +
geom_bar(stat = "identity", position = "stack") +
#geom_text(aes(label = n),
# position = position_stack(vjust = 0.5),
# size = 3.5,
# color = "black",
# check_overlap = TRUE) +
labs(
title = "Political Voice",
x = "Direction of Change",
y = "Percentage",
fill = "ToF System"
)+scale_fill_manual(values = color_palette_18) + # Use the viridis palette
theme_classic() +
theme(
#legend.position = "bottom",
axis.text.x = element_text(angle = 45, hjust = 1) # Rotate x-axis labels for better readability
)
#4. local_recognition
local_recognition = systems_agency_outcomes_long%>%filter(Agency_Social_outcomes == "Local_Recognition")
local_recognition_sum = local_recognition%>%group_by(Direction, TOF_System)%>%count()
local_recognition_totals = local_recognition_sum%>%group_by(Direction)%>%summarise(total_n = sum(n), groups = 'drop')
# Join social equity total counts back to thelocal_recognition sums dataframe
local_recognition_percentage <- local_recognition_sum %>%
left_join(local_recognition_totals, by = "Direction") %>%
mutate(percentage = (n / total_n) * 100)
# Create a stacked bar plot
local_recognition_plot = ggplot(local_recognition_percentage, aes(x = Direction, y = percentage, fill = TOF_System)) +
geom_bar(stat = "identity", position = "stack") +
# geom_text(aes(label = n),
# position = position_stack(vjust = 0.5),
# size = 3.5,
# color = "black",
# check_overlap = TRUE) +
labs(
title = "Local Recognition",
x = "Direction of Change",
y = "Percentage",
fill = "ToF System"
)+scale_fill_manual(values = color_palette_18) + # Use the viridis palette
theme_classic() +
theme(
#legend.position = "bottom",
axis.text.x = element_text(angle = 45, hjust = 1) # Rotate x-axis labels for better readability
)
#5. community_interactions
community_interactions = systems_agency_outcomes_long%>%filter(Agency_Social_outcomes == "Community_Interactions")
community_interactions_sum = community_interactions%>%group_by(Direction, TOF_System)%>%count()
community_interactions_totals = community_interactions_sum%>%group_by(Direction)%>%summarise(total_n = sum(n), groups = 'drop')
# Join social equity total counts back to thecommunity_interactions sums dataframe
community_interactions_percentage <- community_interactions_sum %>%
left_join(community_interactions_totals, by = "Direction") %>%
mutate(percentage = (n / total_n) * 100)
# Create a stacked bar plot
community_interactions_plot = ggplot(community_interactions_percentage, aes(x = Direction, y = percentage, fill = TOF_System)) +
geom_bar(stat = "identity", position = "stack") +
#geom_text(aes(label = n),
# position = position_stack(vjust = 0.5),
# size = 3.5,
# color = "black",
# check_overlap = TRUE) +
labs(
title = "Community Interactions",
x = "Direction of Change",
y = "Percentage",
fill = "ToF System"
)+scale_fill_manual(values = color_palette_18) + # Use the viridis palette
theme_classic() +
theme(
#legend.position = "bottom",
axis.text.x = element_text(angle = 45, hjust = 1) # Rotate x-axis labels for better readability
)
#Plot all together
library(ggpubr)
agency_outcomes_plot = ggarrange(community_interactions_plot, local_recognition_plot, Social_Equity_plot, Trust_plot,
political_voice_plot, ncol = 2, nrow = 3, common.legend = TRUE)
agency_outcomes_plot
#LIMITATIONS OF THE DATASET:CONFLICT OF INTEREST, QUALITY OF STUDY, OTHER DETAILS This chunk of code produces statistics that were used in the Discussion section ‘Limitations of this dataset’ to describe the quality of the studies reviewed.
# What % of the data has an implied conflict?
implied_conflict = coded_data%>%group_by(X1_19)%>%summarise(sum = n())
#approx 15% of the data has implied conflict of interest
# What % of the papers have replicable methos?
replicable_methods = coded_data%>%group_by(X6_4)%>%summarise(sum = n())
#approx 23%% of the data has non replicable methods
# what % of the papers have justified sampling strategy?
sampling_strategy = coded_data%>%group_by(X6_5_a)%>%summarise(sum = n())
# How many of these journals are on Baell's list of predatory journals?
baells_yes = baells%>%filter(Beall_list == "Yes")
baells_yes = baells_yes %>% distinct(X1_6_Journal)
baells_yes$On_List = "On_Bealls_List"
coded_baells = left_join(coded_data, baells_yes, by ="X1_6_Journal")
baells_list_total = coded_baells%>%group_by(On_List)%>%summarise(sum = n())Session information
sessionInfo()## R version 4.1.2 (2021-11-01)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 19045)
##
## Matrix products: default
##
## locale:
## [1] LC_COLLATE=English_India.1252 LC_CTYPE=English_India.1252
## [3] LC_MONETARY=English_India.1252 LC_NUMERIC=C
## [5] LC_TIME=English_India.1252
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] RColorBrewer_1.1-3 ggpubr_0.4.0 knitr_1.40 circlize_0.4.16
## [5] wesanderson_0.3.7 networkD3_0.4 highcharter_0.9.4 lubridate_1.9.3
## [9] forcats_1.0.0 stringr_1.5.1 purrr_1.0.1 readr_2.1.4
## [13] tibble_3.2.1 tidyverse_2.0.0 tidyr_1.3.0 dplyr_1.0.10
## [17] reshape2_1.4.4 ggplot2_3.4.4
##
## loaded via a namespace (and not attached):
## [1] sass_0.4.2 jsonlite_1.8.8 carData_3.0-4
## [4] bslib_0.4.0 assertthat_0.2.1 TTR_0.24.3
## [7] highr_0.11 cellranger_1.1.0 yaml_2.2.1
## [10] pillar_1.9.0 backports_1.2.1 lattice_0.20-45
## [13] glue_1.6.2 rlist_0.4.6.2 digest_0.6.27
## [16] ggsignif_0.6.3 colorspace_2.0-2 cowplot_1.1.1
## [19] htmltools_0.5.5 plyr_1.8.7 pkgconfig_2.0.3
## [22] broom_1.0.1 haven_2.4.1 scales_1.2.1
## [25] openxlsx_4.2.4 rio_0.5.29 tzdb_0.4.0
## [28] timechange_0.2.0 generics_0.1.3 farver_2.1.1
## [31] car_3.0-11 ellipsis_0.3.2 cachem_1.0.6
## [34] withr_2.5.0 cli_3.6.1 quantmod_0.4.26
## [37] readxl_1.4.3 magrittr_2.0.3 evaluate_0.23
## [40] fansi_0.5.0 rstatix_0.7.2 xts_0.12.1
## [43] foreign_0.8-81 tools_4.1.2 data.table_1.14.0
## [46] hms_1.1.2 GlobalOptions_0.1.2 lifecycle_1.0.3
## [49] munsell_0.5.0 zip_2.2.0 compiler_4.1.2
## [52] jquerylib_0.1.4 tinytex_0.41 rlang_1.1.0
## [55] grid_4.1.2 rstudioapi_0.14 htmlwidgets_1.6.1
## [58] igraph_1.3.5 labeling_0.4.2 rmarkdown_2.16
## [61] gtable_0.3.1 abind_1.4-5 DBI_1.1.3
## [64] curl_4.3.2 R6_2.5.1 gridExtra_2.3
## [67] zoo_1.8-11 fastmap_1.1.0 utf8_1.2.1
## [70] shape_1.4.6 stringi_1.7.6 Rcpp_1.0.7
## [73] vctrs_0.6.1 tidyselect_1.2.0 xfun_0.33