First Insight
Describe the Annual per capita poverty threshold. compare this threshold across different regions, province and District and across time.
TOTAL_APCPT_YEAR %>%
ggplot(aes( x= as.Date(paste0(Year, "-01-01")), y = `Total Annual Per Capita Poverty Threshold (in Pesos)`)) +
geom_line(size = 0.8, colour = "#112446") +
geom_point( shape=21, color="black", fill="#69b3a2", size=6)+
geom_smooth(method = lm, se = F) +
annotate("text", x = as.Date("2010-01-01"), y = 2.2e6, label = paste0("Correlation = ", round(cor(TOTAL_APCPT_YEAR$Year, TOTAL_APCPT_YEAR$`Total Annual Per Capita Poverty Threshold (in Pesos)`, use = "complete.obs"), 5))) +
labs(y = "Pesos",
x = "Year",
title = "Total Annual Per Capita Poverty Threshold (in Pesos) Per Year") +
theme_classic() +
scale_x_date(date_breaks = "1 year",
date_labels = "%Y",
limits = as.Date(c("2006-01-01", "2015-01-01")))
## `geom_smooth()` using formula 'y ~ x'
Graph 1.0 explanation
The line graph shows the Annual Per Capita Poverty Threshold over a period of 10 years, from 2006 to 2015. We can see that there is a sharp rise in the poverty threshold during this period.
The line on the graph appears to be almost straight, indicating that the increase in poverty threshold was steady and linear during these 10 years.
Moreover, if we look at the correlation coefficient (a measure of how closely two variables are related), it has a value of 0.9956, which is very close to 1. This indicates a strong positive correlation between the year and the Annual Per Capita Poverty Threshold, medianing that as the year increases, the poverty threshold also increases.
Another interesting observation is that there is a higher increase in the poverty threshold every 3 years from 2006 to 2015. This could be due to various factors such as inflation, economic growth, and government policies.
Overall, the graph clearly illustrates the increase in poverty threshold over time, and the strong correlation between the year and the poverty threshold.
Region
region_total <- wide_df %>%
group_by(region) %>%
summarize(total_Poverty_Threshold = sum(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(total_Poverty_Threshold))
region_total
## # A tibble: 17 x 2
## region total_Poverty_Threshold
## <chr> <dbl>
## 1 Region III 522265
## 2 CAR 433308
## 3 Region V 417982
## 4 Region VI 415185
## 5 Region VIII 411670
## 6 Region II 378401
## 7 Region VI-A 364812
## 8 Region X 353059
## 9 ARMM 346354
## 10 Region IV-B 329981
## 11 NCR 321108
## 12 CARAGA 298774
## 13 Region XI 293953
## 14 Region I 287347
## 15 Region VII 280534
## 16 Region XII 276311
## 17 Region IX 203423
region_total %>%
mutate(region = fct_reorder(region, total_Poverty_Threshold, .desc = F)) %>%
ggplot(data = , aes(x =region, y = total_Poverty_Threshold)) +
geom_segment(aes(x = region, y = 0, xend = region, yend = total_Poverty_Threshold), color = "gray") +
geom_point(size = 3, color = "#3960E9", shape = 21, fill = "white", stroke = 1.2) +
coord_flip() +
labs(x = "Region", y = "Total Annual Per Capita Poverty Threshold (in Pesos)", title = "Total Annual Per Capita Poverty Threshold by Region (2006-2015)") +
theme_minimal() +
scale_y_continuous(labels = comma)
Graph 1.1 explanation
The lollipop chart above shows the total annual per capita poverty threshold by region in the Philippines, ranked from highest to lowest. The ranking begins with Region III having the highest threshold and Region IX having the lowest. It is interesting to note that the ranking is not necessarily consistent with the geographical location of the regions.
Regions III, CAR, V, VI, and VIII have the highest poverty threshold, while Regions IV-B, NCR, CARAGA, XI, and I have a lower threshold. Region IX has the lowest poverty threshold. This ranking can provide insight into the economic conditions of each region and can be useful for policy makers in allocating resources for poverty reduction programs.
It is important to note that poverty threshold is not the same as poverty incidence. Poverty threshold refers to the minimum income required to meet basic needs such as food, clothing, and shelter, while poverty incidence refers to the percentage of individuals or families falling below the poverty line.
Overall, the graph highlights the need for targeted poverty reduction efforts in regions with the lowest poverty threshold, as well as continued support for poverty reduction programs in regions with high poverty threshold to maintain their progress.
region_total <- wide_df %>%
group_by(region, Year) %>%
summarize(total_Poverty_Threshold = sum(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(total_Poverty_Threshold))
## `summarise()` has grouped output by 'region'. You can override using the
## `.groups` argument.
region_total
## # A tibble: 68 x 3
## # Groups: region [17]
## region Year total_Poverty_Threshold
## <chr> <dbl> <dbl>
## 1 Region III 2015 162575
## 2 Region III 2012 138464
## 3 CAR 2015 130509
## 4 Region VIII 2015 129732
## 5 Region V 2015 128877
## 6 Region VI 2015 128632
## 7 Region III 2009 123194
## 8 CAR 2012 116271
## 9 Region II 2015 115874
## 10 Region VI-A 2015 113132
## # ... with 58 more rows
top_region_per_year <- function(year){
region_total <- wide_df %>%
filter(Year == year) %>%
group_by(region) %>%
summarize(total_Poverty_Threshold = sum(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(total_Poverty_Threshold))
region_total %>%
mutate(region = fct_reorder(region, total_Poverty_Threshold, .desc = F)) %>%
ggplot(data = , aes(x =region, y = total_Poverty_Threshold)) +
geom_segment(aes(x = region, y = 0, xend = region, yend = total_Poverty_Threshold),
color = "gray") +
geom_point(size = 3, color = "#3960E9", shape = 21, fill = "white", stroke = 1.2) +
coord_flip() +
labs(x = "", y = "",
title = year) +
theme_minimal() +
scale_y_continuous(labels = comma)
}
top_region_2006 <- top_region_per_year("2006")
top_region_2009 <- top_region_per_year("2009")
top_region_2012 <- top_region_per_year("2012")
top_region_2015 <- top_region_per_year("2015")
(top_region_2006 | top_region_2009) / (top_region_2012 | top_region_2015) +
plot_annotation(title = "Regions Total Annual Per Capita Poverty Threshold per Year",
theme = theme(plot.title = element_text(hjust = 0.5, face = "bold")))
Graph 1.2 Explanation
- The graph displays the total annual per capita poverty threshold by region from 2006 to 2015. It shows that Regions III and CAR consistently hold the top 2 spots with the highest poverty threshold, while Region IX consistently holds the last spot with the lowest poverty threshold. This indicates a significant difference in poverty levels across regions, highlighting the need for targeted and region-specific poverty reduction strategies. Understanding the variation in poverty threshold across regions is essential in designing policies and interventions that effectively address the specific needs and circumstances of each region.
Province
wide_df %>%
filter(adm_level == "region")
## # A tibble: 0 x 12
## # ... with 12 variables: Year <dbl>, place <chr>, adm_level <chr>,
## # region <chr>, unit <chr>,
## # Annual Per Capita Poverty Threshold (in Pesos) <dbl>,
## # Poverty Incidence among Families (%) <dbl>,
## # Magnitude of Poor Families <dbl>,
## # Poverty Incidence among Population (%) <dbl>,
## # Magnitude of Poor Population <dbl>, ...
# Filter the top 10 provinces with the highest median Annual Per Capita Poverty Threshold (in Pesos)
top10 <- wide_df %>%
filter(adm_level == "Province") %>%
group_by(place) %>%
summarize(median_Poverty_Threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
top_n(10, median_Poverty_Threshold) %>%
arrange(desc(median_Poverty_Threshold))
# Sort the wide_df dataset by median poverty threshold and reorder place factor variable
sorted_df <- wide_df %>%
filter(adm_level == "Province") %>%
group_by(place) %>%
summarize(median_poverty_threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(median_poverty_threshold)) %>%
inner_join(wide_df, by = "place") %>%
mutate(place = fct_reorder(place, median_poverty_threshold))
# Plot the top 10 provinces using a horizontal box plot
sorted_df %>%
filter(place %in% top10$place) %>%
ggplot(aes(y = `Annual Per Capita Poverty Threshold (in Pesos)`, x = place, fill )) +
geom_boxplot(fill = "#69b3a2") +
labs(title = "Top 10 Provinces with the Highest median Annual Per Capita Poverty Threshold",
y = "Annual Per Capita Poverty Threshold (in Pesos)", x = "") +
theme_minimal() +
coord_flip() +
theme(plot.title = element_text(hjust = 0.5, face = "bold"))
Graph 1.3 explanation
The boxplot above shows the distribution of the Annual Per Capita Poverty Threshold (in Pesos) per Province. There are a total of 79 provinces in the Philippines, and based on the graph, we can identify the Top 10 Provinces with the Highest median Annual Per Capita Poverty Threshold. These provinces are Batanes (with a median poverty threshold of 23,234.50 pesos), Zambales (20,166.50 pesos), Nueva Ecija (19,789.50 pesos), Mt. Province (19,628.00 pesos), Agusan del Sur (19,370.50 pesos), Compostela Valley(19,360.00 pesos), Rizal (19,351.50 pesos) and Bulacan (19,172.00 pesos). .
The highest median Annual Per Capita Poverty Threshold is found in Batanes, which also has the highest median value, indicating that this province is relatively higher cost of meeting basic needs compared to all other provinces in the Philippines. Furthermore, the Annual Per Capita Poverty Threshold in Batanes is the highest among all provinces.
In top provinces also We can also observe that the IQR is wider in Lanao del Sur, which suggests that the poverty threshold is more widely spread out in this province compared to others. In contrast, the smallest spread of IQR is found in the province of Agusan del Sur.
top10_total_Poverty_Threshold <- wide_df %>%
filter(adm_level == "Province") %>%
group_by(place, Year) %>%
summarize(total_Poverty_Threshold = sum(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(total_Poverty_Threshold)) %>%
group_by(Year) %>%
top_n(10, total_Poverty_Threshold)
## `summarise()` has grouped output by 'place'. You can override using the
## `.groups` argument.
top10_total_Poverty_Threshold
## # A tibble: 40 x 3
## # Groups: Year [4]
## place Year total_Poverty_Threshold
## <chr> <dbl> <dbl>
## 1 Batanes 2015 29118
## 2 Zambales 2015 26473
## 3 Cavite 2015 24882
## 4 Bataan 2015 24770
## 5 Batanes 2012 24693
## 6 Davao del Norte 2015 24424
## 7 Rizal 2015 24198
## 8 Bukidnon 2015 23682
## 9 Mt. Province 2015 23620
## 10 Nueva Ecija 2015 23403
## # ... with 30 more rows
top10_per_year <- function(year){
top10_total_Poverty_Threshold %>%
filter(Year == year) %>%
mutate(place = fct_reorder(place, total_Poverty_Threshold, .desc = F)) %>%
ggplot(aes(x = place, weight = total_Poverty_Threshold, fill = total_Poverty_Threshold))+
geom_bar(width = 0.6, show.legend = F) +
coord_flip() +
labs(x = "", y = "", title = year) +
theme_minimal() +
theme(
plot.title = element_text(hjust = .5, size = 10)
) +
scale_fill_gradient(low = "#50D4C5", high = "#3960E9") +
geom_text(aes(label = total_Poverty_Threshold, y = total_Poverty_Threshold + 50), color = "black", size = 3, hjust = 1.1)
}
top10_2006 <- top10_per_year("2006")
top10_2009 <- top10_per_year("2009")
top10_2012 <- top10_per_year("2012")
top10_2015 <- top10_per_year("2015")
(top10_2006 | top10_2009) / (top10_2012 | top10_2015) +
plot_annotation(title = "Top 10 Provinces with the Highest Total Annual Per Capita Poverty Threshold per Year",
theme = theme(plot.title = element_text(hjust = 0.5, face = "bold")))
Graph 1.4 explanation
The graph above shows the top 10 provinces with the highest Annual Per Capita Poverty Threshold per Year from 2000 to 2015. We can see that over the years, the value of the poverty threshold for each province has been increasing, and this trend is consistent across all provinces.
One province that stands out is Batanes, which consistently has a high Annual Per Capita Poverty Threshold over the years. It also has a higher average Annual Per Capita Poverty Threshold compared to other provinces.
The provinces with the highest median Annual Per Capita Poverty Threshold - Batanes, Nueva Ecija, Zambales, and Mt. Provinces - are consistently seen in the top 10 provinces with the highest total Capita Poverty Threshold per year from 2006 to 2015. This suggests that these findings are related to each other.
In summary, the graph shows that the poverty threshold has been increasing over the years for all provinces, and Batanes consistently has a higher poverty threshold compared to other provinces. The top provinces with the highest median Annual Per Capita Poverty Threshold are also seen in the top 10 provinces with the highest total Capita Poverty Threshold per year, indicating a relationship between the two findings.
wide_df %>%
filter(adm_level == "Province") %>%
group_by(place) %>%
summarize(median_Poverty_Threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
top_n(10, -median_Poverty_Threshold) %>%
arrange(median_Poverty_Threshold)
## # A tibble: 10 x 2
## place median_Poverty_Threshold
## <chr> <dbl>
## 1 Western Samar 15491
## 2 Palawan 15557
## 3 Tawi-tawi 15622.
## 4 Negros Occidental 16082
## 5 Negros Oriental 16107
## 6 Zamboanga del Sur 16320
## 7 Biliran 16372.
## 8 Marinduque 16429
## 9 North Cotabato 16601
## 10 Kalinga 16640.
# Filter the top 10 provinces with the lowest median Annual Per Capita Poverty Threshold (in Pesos)
bottom10 <-wide_df %>%
filter(adm_level == "Province") %>%
group_by(place) %>%
summarize(median_Poverty_Threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
top_n(10, -median_Poverty_Threshold) %>%
arrange(median_Poverty_Threshold)
# Sort the wide_df dataset by median poverty threshold and reorder place factor variable
sorted_df <- wide_df %>%
filter(adm_level == "Province") %>%
group_by(place) %>%
summarize(median_poverty_threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(median_poverty_threshold) %>%
inner_join(wide_df, by = "place") %>%
mutate(place = fct_reorder(place, median_poverty_threshold, .desc = T))
# Plot the top 10 provinces using a horizontal box plot
sorted_df %>%
filter(place %in% bottom10$place) %>%
ggplot(aes(y = `Annual Per Capita Poverty Threshold (in Pesos)`, x = place, fill )) +
geom_boxplot(fill = "#69b3a2") +
labs(title = "Top 10 Provinces with the lowest median Annual Per Capita Poverty Threshold",
y = "Annual Per Capita Poverty Threshold (in Pesos)", x = "") +
theme_minimal() +
coord_flip() +
theme(plot.title = element_text(hjust = 0.5, face = "bold"))
Graph 1.5 explanation
- The boxplot above displays the distribution of Annual Per Capita Poverty Threshold (in Pesos) across all provinces in the Philippines. From the boxplot, we can see that the province of Western Samar has the lowest median value of Annual Per Capita Poverty Threshold (in Pesos) compared to other provinces.
top10_total_Poverty_Threshold_lowest <- wide_df %>%
filter(adm_level == "Province") %>%
group_by(place, Year) %>%
summarize(total_Poverty_Threshold = sum(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(total_Poverty_Threshold)) %>%
group_by(Year) %>%
top_n(10, - total_Poverty_Threshold) %>%
arrange(total_Poverty_Threshold)
## `summarise()` has grouped output by 'place'. You can override using the
## `.groups` argument.
top10_total_Poverty_Threshold_lowest
## # A tibble: 40 x 3
## # Groups: Year [4]
## place Year total_Poverty_Threshold
## <chr> <dbl> <dbl>
## 1 Zamboanga del Sur 2006 11508
## 2 Negros Oriental 2006 11518
## 3 Palawan 2006 11521
## 4 Tawi-tawi 2006 11557
## 5 Western Samar 2006 11607
## 6 Negros Occidental 2006 11826
## 7 Biliran 2006 11841
## 8 Aurora 2006 11883
## 9 North Cotabato 2006 12077
## 10 Camiguin 2006 12152
## # ... with 30 more rows
top10_per_year <- function(year){
top10_total_Poverty_Threshold_lowest %>%
filter(Year == year) %>%
mutate(place = fct_reorder(place, total_Poverty_Threshold, .desc = F)) %>%
ggplot(aes(x = place, weight = total_Poverty_Threshold, fill = total_Poverty_Threshold))+
geom_bar(width = 0.6, show.legend = F) +
coord_flip() +
labs(x = "", y = "", title = year) +
theme_minimal() +
theme(
plot.title = element_text(hjust = .5, size = 10)
) +
scale_fill_gradient(low = "#50D4C5", high = "#3960E9") +
geom_text(aes(label = total_Poverty_Threshold, y = total_Poverty_Threshold + 50), color = "black", size = 3, hjust = 1.1)
}
top10_2006 <- top10_per_year("2006")
top10_2009 <- top10_per_year("2009")
top10_2012 <- top10_per_year("2012")
top10_2015 <- top10_per_year("2015")
(top10_2006 | top10_2009) / (top10_2012 | top10_2015) +
plot_annotation(title = "Top 10 Provinces with the Lowest Total Annual Per Capita Poverty Threshold per Year",
theme = theme(plot.title = element_text(hjust = 0.5, face = "bold")))
Graph 1.6 explanation
The graph above displays the Top 10 Provinces with the Lowest Total Annual Per Capita Poverty Threshold per Year in the Philippines. From the graph, we can see that the provinces with the lowest total values are Zamboanga del Sur, Negros Oriental, Western Samar, and Tawi-Tawi in years 2006, 2009, 2012, and 2015, respectively. We can also observe that the total values of these provinces have been increasing over the years.
It is surprising to note that the smallest province, Western Samar, did not consistently belong to the top in terms of having the lowest Total Annual Per Capita Poverty Threshold per Year, unlike the province of Batanes in an earlier analysis where it had a higher mean and also led in having the highest Total Annual Per Capita Poverty Threshold over the years.
Moreover, the provinces of Negros Oriental, Palawan, Western Samar, and Tawi-Tawi consistently belong to the top provinces with the lowest median Total Annual Per Capita Poverty Threshold. We can also observe that some provinces with the lowest median values can be seen in some years, indicating the accuracy of our analysis as the results of the graph are related to the previous analysis.
In summary, the graph shows that the total Annual Per Capita Poverty Threshold values of the provinces with the lowest values have been increasing over the years. The smallest province, Western Samar, did not consistently belong to the top, while other provinces like Negros Oriental, Palawan, Western Samar, and Tawi-Tawi consistently belong to the top provinces with the lowest median Total Annual Per Capita Poverty Threshold. The results of the graph are related to the previous analysis, indicating the accuracy of our analysis.
District
top10 <- wide_df %>%
filter(adm_level == "District") %>%
group_by(place) %>%
summarize(median_Poverty_Threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
top_n(10, median_Poverty_Threshold) %>%
arrange(desc(median_Poverty_Threshold))
top10
## # A tibble: 4 x 2
## place median_Poverty_Threshold
## <chr> <dbl>
## 1 1st District 19786.
## 2 2nd District 19786.
## 3 3rd District 19786.
## 4 4th District 19786.
# Filter the top district with the lowest median Annual Per Capita Poverty Threshold (in Pesos)
bottom10 <-wide_df %>%
filter(adm_level == "District") %>%
group_by(place) %>%
summarize(median_Poverty_Threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
top_n(10, -median_Poverty_Threshold) %>%
arrange(desc(median_Poverty_Threshold))
# Sort the wide_df dataset by median poverty threshold and reorder place factor variable
sorted_df <- wide_df %>%
filter(adm_level == "District") %>%
group_by(place) %>%
summarize(median_poverty_threshold = median(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(median_poverty_threshold)) %>%
inner_join(wide_df, by = "place") %>%
mutate(place = fct_reorder(place, median_poverty_threshold))
# Plot the top district using a horizontal box plot
sorted_df %>%
filter(place %in% bottom10$place) %>%
ggplot(aes(y = `Annual Per Capita Poverty Threshold (in Pesos)`, x = place, fill )) +
geom_boxplot(fill = "#69b3a2", width = .5) +
labs(title = "Districts in NCR with the median Annual Per Capita Poverty Threshold",
y = "Annual Per Capita Poverty Threshold (in Pesos)", x = "") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5, face = "bold"))
Graph 1.7 explanation
For the distribution in District in adml_level is almost same value of 19785.5 pesos
The
top10_total_Poverty_Threshold_district <- wide_df %>%
filter(adm_level == "District") %>%
group_by(place, Year) %>%
summarize(total_Poverty_Threshold = sum(`Annual Per Capita Poverty Threshold (in Pesos)`, na.rm = TRUE)) %>%
arrange(desc(total_Poverty_Threshold)) %>%
group_by(Year) %>%
top_n(10, total_Poverty_Threshold)
## `summarise()` has grouped output by 'place'. You can override using the
## `.groups` argument.
top10_total_Poverty_Threshold
## # A tibble: 40 x 3
## # Groups: Year [4]
## place Year total_Poverty_Threshold
## <chr> <dbl> <dbl>
## 1 Batanes 2015 29118
## 2 Zambales 2015 26473
## 3 Cavite 2015 24882
## 4 Bataan 2015 24770
## 5 Batanes 2012 24693
## 6 Davao del Norte 2015 24424
## 7 Rizal 2015 24198
## 8 Bukidnon 2015 23682
## 9 Mt. Province 2015 23620
## 10 Nueva Ecija 2015 23403
## # ... with 30 more rows
top10_per_year_district <- function(year){
top10_total_Poverty_Threshold_district %>%
filter(Year == year) %>%
mutate(place = fct_reorder(place, total_Poverty_Threshold, .desc = F)) %>%
ggplot(aes(x = place, weight = total_Poverty_Threshold, fill = total_Poverty_Threshold))+
geom_bar(width = 0.6, show.legend = F) +
coord_flip() +
labs(x = "", y = "", title = year) +
theme_minimal() +
theme(
plot.title = element_text(hjust = .5, size = 10)
) +
scale_fill_gradient(low = "#50D4C5", high = "#3960E9") +
geom_text(aes(label = total_Poverty_Threshold, y = total_Poverty_Threshold + 50), color = "black", size = 3, hjust = 1.1)
}
top10_2006 <- top10_per_year_district("2006")
top10_2009 <- top10_per_year_district("2009")
top10_2012 <- top10_per_year_district("2012")
top10_2015 <- top10_per_year_district("2015")
(top10_2006 | top10_2009) / (top10_2012 | top10_2015) +
plot_annotation(title = "Districts in NCR with the Highest Total Annual Per Capita Poverty Threshold per Year",
theme = theme(plot.title = element_text(hjust = 0.5, face = "bold")))
Graph 1.8 explanation
The bar graph of districts in NCR with the Highest Total Annual Per Capita Poverty Threshold per Year poverty threshold per year has the same value per year as it has no insight at all.
Second Insight
Compare the Total Magnitude of Poor Families from 1991 – 2015 and per region.
TOTAL_MPF_YEAR <- wide_df %>%
group_by(Year) %>%
summarise("Total Magnitude of Poor Families" = sum(`Magnitude of Poor Families`, na.rm = TRUE)) %>%
arrange(desc(`Total Magnitude of Poor Families`))
TOTAL_MPF_YEAR
## # A tibble: 4 x 2
## Year `Total Magnitude of Poor Families`
## <dbl> <dbl>
## 1 2012 4189937
## 2 2009 4016936
## 3 2006 3791135
## 4 2015 3729660
TOTAL_MPF_YEAR %>%
ggplot(aes(x = as.Date(paste0(Year, "-01-01")), y = `Total Magnitude of Poor Families`)) +
geom_line(size = 0.8, colour = "#112446") +
geom_point(shape = 21, color = "black", fill = "#69b3a2", size = 6) +
labs(y = "Famillies",
x = "Year",
title = "Total Magnitude of Poor Families Per Year") +
theme_classic() +
scale_x_date(date_breaks = "1 year",
date_labels = "%Y",
limits = as.Date(c("2006-01-01", "2015-01-01"))) +
scale_y_continuous(limits = c(2000000, NA))+
theme(plot.title = element_text(hjust = 0.5, face = "bold"))
Graph 2.0 explanation
The line graph above displays the total magnitude of poor families per year in a given population from 2006 to 2015. We can see that there was an increase of around 100,000 to 300,000 total families falling under the poverty line over the course of these years. This indicates a concerning trend of growing poverty within the population during this time period.
However, it is worth noting that from the year 2012 to 2015, there is a decrease of 460,277 total families. This is an encouraging trend, as it suggests that measures taken to combat poverty may have started to take effect, resulting in a decline in the total number of families falling under the poverty line.
It is important to keep in mind that the line graph does not show a consistent increase or decrease in the total magnitude of poor families per year, making it difficult to identify a clear trend. Nevertheless, the graph does provide insight into the changes in poverty rates over time and can be used to identify any significant changes that have occurred.
Overall, the line graph highlights the need for continued efforts to combat poverty in the population, while also acknowledging that progress can be made in reducing the total number of families falling under the poverty line.
TOTAL_MPF_region <- wide_df %>%
group_by(region) %>%
summarise("Total Magnitude of Poor Families" = sum(`Magnitude of Poor Families`, na.rm = TRUE)) %>%
arrange(desc(`Total Magnitude of Poor Families`))
TOTAL_MPF_region
## # A tibble: 17 x 2
## region `Total Magnitude of Poor Families`
## <chr> <dbl>
## 1 Region VII 1589681
## 2 Region V 1470263
## 3 Region VI 1316966
## 4 Region VIII 1202323
## 5 Region X 1194120
## 6 Region XII 1151583
## 7 Region IX 994771
## 8 ARMM 986681
## 9 Region XI 943357
## 10 Region VI-A 904149
## 11 Region III 903259
## 12 CARAGA 766449
## 13 Region I 630997
## 14 Region IV-B 608278
## 15 Region II 512696
## 16 NCR 294103
## 17 CAR 257992
TOTAL_MPF_region %>%
mutate(region = fct_reorder(region, `Total Magnitude of Poor Families`, .desc = F)) %>%
ggplot(aes(x = region, weight = `Total Magnitude of Poor Families`, fill = `Total Magnitude of Poor Families`))+
geom_bar(width = 0.8, show.legend = F) +
coord_flip() +
labs(x = "", y = "", title = "Total Magnitude of Poor Families Per Region") +
theme_minimal() +
theme(
plot.title = element_text(hjust = .5, size = 10)
) +
scale_fill_gradient(low = "#50D4C5", high = "#3960E9") +
geom_text(aes(label = scales::comma_format()(`Total Magnitude of Poor Families`), y = `Total Magnitude of Poor Families` + 50), color = "black", size = 3, hjust = 1.1) +
theme(plot.title = element_text(hjust = 0.5, face = "bold"))
graph 2.1 explanation
- The bar graph displays the total magnitude of poor families per Region in the Philippines. Region VII has the highest magnitude with 1,589,681 poor families, followed by Region V and VI. On the other hand, CAR has the lowest magnitude of poor families, followed by NCR and Region II. The graph provides insight into the distribution of poverty across the regions and can be used to inform policies and interventions aimed at reducing poverty in the country.
Third Insight
Correlations between different poverty indicators: You can also explore the relationships between different poverty indicators to understand how poverty incidence, magnitude, and subsistence incidence are related
selected_vars <- wide_df %>%
select(`Annual Per Capita Poverty Threshold (in Pesos)`, `Poverty Incidence among Families (%)`, `Magnitude of Poor Families`, `Poverty Incidence among Population (%)`,`Magnitude of Poor Population`,`Subsistence Incidence among Population (%)`, `Magnitude of Subsistence Poor Population`)
# Compute the correlation matrix using the cor() function
cor_matrix <- cor(selected_vars, use = "pairwise.complete.obs")
# Convert the correlation matrix to a data frame
cor_df <- as.data.frame(cor_matrix)
# Add a column with the variable names
cor_df$variable1 <- rownames(cor_df)
# Convert the data frame to a tidy format using the pivot_longer() function
cor_tidy <- pivot_longer(cor_df, -variable1, names_to = "variable2", values_to = "correlation")
# Normalize the correlation values to a continuous scale between 0 and 1
cor_tidy$correlation_norm <- (cor_tidy$correlation + 1) / 2
# Plot the correlation matrix as a heatmap using ggplot2
ggplot(cor_tidy, aes(variable1, variable2, fill = correlation_norm)) +
geom_tile() +
scale_fill_gradient2(low="white", mid = "#50D4C5", high="blue", midpoint = 0.5, na.value = "white") +
theme_minimal() +
labs(x = NULL, y = NULL, fill = "Correlation", title = "poverty indicators Correlation") +
theme(axis.text.x = element_text( angle = 90)) +
geom_text(aes(label = round(correlation,2)), size = 3, color = "white")
Graph 3.0 explanation
The matrix shows the correlation coefficients between poverty indicators in the Philippines. The coefficients range from -1 to 1, with positive values indicating a positive correlation and negative values indicating a negative correlation.
This matrix illustrates the degree to which poverty indicators in the Philippines are correlated with each other. The high correlations suggest that addressing one aspect of poverty can have spillover effects on other indicators, underscoring the importance of a coordinated and comprehensive approach.
The values indicate a high positive correlation between the magnitude of subsistence poor population and the magnitude of poor population at 0.93. Additionally, there is a high positive correlation between poverty incidence among the population and poverty incidence among families at 0.99. Furthermore, poverty incidence among families and subsistence incidence among the population have a high positive correlation of 0.92. These findings suggest that these poverty indicators are closely related and can provide a comprehensive understanding of poverty in a population.