Assignment 2.qmd

---
title: "Assignment 2"
author: "Jazmin Hernandez"
format: html
editor: visual
embed-resources: true 
---

# **Assignment 02 - Data Viz and Wrangling**

## Data Wrangling

```{r}
library(dplyr)
library(readr)
individual_data <- "C:/Users/jhern/Downloads/chs_individual.csv"
regional_data <- "C:/Users/jhern/Downloads/chs_regional.csv"
individual_data <- read_csv(individual_data)
regional_data <- read_csv(regional_data)
```

```{r}
merged_data <- left_join(individual_data, regional_data, by = "townname")
head(merged_data)
```

```{r}
row_count <- nrow(merged_data)
cat("Rows in merged_data:", row_count, "\n")
duplicate_count <- sum(duplicated(merged_data))
cat("Number of duplicate rows in merged_data:", duplicate_count, "\n")
```

```{r}
Mode <- function(x) {
  x <- x[!is.na(x)]  
  uniq_x <- unique(x)  
  if (length(uniq_x) == 0) return(NA) 
  uniq_x[which.max(tabulate(match(x, uniq_x)))]  
}
```

```{r}
merged_data |>
group_by(male, hispanic) |>
mutate(
across(where(is.numeric), 
           ~ if_else(is.na(.), mean(., na.rm = TRUE), .)),
across(where(is.character), 
           ~ if_else(is.na(.), Mode(.), .))                 
  ) |> 
  ungroup() 
```

```{r}
missing_values_count <- colSums(is.na(merged_data))
print(missing_values_count)
```

```{r}
merged_data <- merged_data |>
mutate(obesity_level = case_when(
    bmi < 14 ~ "Underweight",
    bmi >= 14 & bmi <= 22 ~ "Normal",
    bmi > 22 & bmi <= 24 ~ "Overweight",
    bmi > 24 ~ "Obese",
    TRUE ~ NA_character_  
  ))
```

```{r}
summary_table <- merged_data |>
group_by(obesity_level) |>
summarize(
min_bmi = min(bmi, na.rm = TRUE),
max_bmi = max(bmi, na.rm = TRUE),
total_observations = n(),
.groups = 'drop'  
  )
print(summary_table)
```

```{r}
merged_data <- merged_data|>
  mutate(smoke_gas_exposure = case_when(
    smoke == 0 & gasstove == 0 ~ "Non-exposed",
    smoke == 1 & gasstove == 0 ~ "Second Hand Smoke Only",
    smoke == 0 & gasstove == 1 ~ "Gas Stove Only",
    smoke == 1 & gasstove == 1 ~ "Both Exposures",
    TRUE ~ NA_character_  
  ))
head(merged_data)
```

```{r}
summary_town <- merged_data |>
group_by(townname) |>
summarize(
average_fev = mean(fev, na.rm = TRUE),
sd_fev = sd(fev, na.rm = TRUE),
.groups = 'drop'
  )
print(summary_town)
```

```{r}
summary_sex <- merged_data |>
group_by(male) |>
summarize(
average_fev = mean(fev, na.rm = TRUE),
sd_FEV1 = sd(fev, na.rm = TRUE),
.groups = 'drop'
  )
print(summary_sex)
```

```{r}
summary_obesity <- merged_data |>
group_by(obesity_level) |>
summarize(
average_fev = mean(fev, na.rm = TRUE),
sd_fev = sd(fev, na.rm = TRUE),
.groups = 'drop'
  )
print(summary_obesity)
```

```{r}
summary_exposure <- merged_data |>
group_by(smoke_gas_exposure) |>
summarize(
average_fev = mean(fev, na.rm = TRUE),
sd_fev = sd(fev, na.rm = TRUE),
.groups = 'drop'
  )
print(summary_exposure)
```

## Looking at the Data (EDA)

1.  What is the association between BMI and FEV (forced expiratory volume)? From the scatter plot of BMI vs FEV, there does appear to be a positive correlation where as BMI increases, so does forced expiratory volume.

```{r}
colSums(is.na(merged_data|> select(bmi, fev)))
merged_data <- na.omit(merged_data)
```

```{r}
layout(matrix(1:2, nrow=1))
hist(merged_data$bmi)
hist(merged_data$fev)
```

```{r}
library(ggplot2)
ggplot(merged_data, aes(x = bmi, y = fev)) + 
  geom_point(alpha = 0.5) + 
  geom_smooth(method = "lm", color = "purple") + 
  labs(title = "Scatter Plot: BMI vs FEV", x = "BMI", y = "FEV") + 
  theme_minimal()
```

2.  What is the association between smoke and gas exposure and FEV? From the box plot, there does not appear to be an association between smoke and gas exposure and forced expiratory volume. This is because of the considerable overlapping among all four categories smoke and gas exposure.

```{r}
colSums(is.na(merged_data|> select(smoke_gas_exposure, fev)))
merged_data <- na.omit(merged_data)
```

```{r}
ggplot(merged_data, aes(x = factor(smoke_gas_exposure), y = fev)) + 
geom_boxplot(fill = "pink") + 
labs(title = "Boxplot of FEV by Smoke and Gas Exposure", 
x = "Smoke and Gas Exposure (0 = No, 1 = Yes)", y = "FEV") + 
theme_minimal()
```

3.  What is the association between PM2.5 exposure and FEV? From the scatterplot, there appears to be no association between PM 2.5 exposure and forced expiratory exposure as seen by the linear model only having a very slight decrease from 10 to about 33 pm2.5 exposure.

```{r}
ggplot(merged_data, aes(x = pm2_5_fr, y = fev)) +
  geom_point() +
  geom_smooth(method = "lm", se = FALSE, color = "green") +
  labs(title = "Scatter plot of PM2.5 vs FEV", x = "PM2.5 Exposure", y = "FEV") +
  theme_minimal()
```

## Visualization

1.  Facet plot showing scatterplots with regression lines of BMI vs FEV by “townname”. All scatterplots do show an association between BMI and FEV. We can see that throughout all towns, as BMI increases, so does forced expiratory volume. However, in areas such as Riverside and Alpine, this association seems more gradual.

```{r}
ggplot(merged_data, aes(x = bmi, y = fev)) +
geom_point(alpha = 0.6) +  
geom_smooth(method = "lm", se = FALSE, color = "lightblue") + 
facet_wrap(~ townname) +  
labs(
    title = "BMI vs FEV by Town",
    x = "BMI",
    y = "FEV",
  ) +
  theme_minimal()
```

2.  Stacked histograms of FEV by BMI category and FEV by smoke/gas exposure. Use different color schemes than the ggplot default.

The BMI categories seem to fall within similar ranges in this stacked histogram. Being underweight appears to have the lowest counts of FEV and being obese has the highest counts.

```{r}
ggplot(merged_data, aes(x = fev, fill = obesity_level)) +
  geom_histogram(position = "stack", bins = 15, alpha = 0.7) +
  scale_fill_manual(values = c("Underweight" = "darkred", 
                                "Normal weight" = "lavender", 
                                "Overweight" = "lightgreen", 
                                "Obese" = "darkblue")) +
  labs(
    title = "Stacked Histogram of FEV by BMI Category",
    x = "Forced Expiratory Volume (FEV)",
    y = "Count",
    fill = "BMI Category"
  ) +
  theme_minimal()
```

The stacked histogram of FEV by smoke/gas exposure shows the category for both exposures has the highest count with second-hand smoke category having the lowest count.

```{r}
ggplot(merged_data, aes(x = fev, fill = smoke_gas_exposure)) +
geom_histogram(position = "stack", bins = 30, alpha = 0.7) +
scale_fill_manual(values = c("purple", "yellow", "blue", "red")) +
labs(
title = "Stacked Histogram of FEV by Smoke/Gas Exposure",
x = "FEV",
y = "Count"
  ) +
theme_minimal()
```

3.  Barchart of BMI by smoke/gas exposure. Taking the average BMI by smoke/gas exposure category we can see that there does not seem to be a significant association between smoke/gas exposure and BMI. Gas stove exposure and non-exposed are the two categories that seem to be slightly associated with a lower BMI while having both exposures and secondhand smoke only seem to be slightly more associated with higher BMI.

```{r}
bmi_summary <- merged_data |>
  group_by(smoke_gas_exposure) |>
  summarize(avg_bmi = mean(bmi, na.rm = TRUE))

ggplot(bmi_summary, aes(x = smoke_gas_exposure, y = avg_bmi, fill = smoke_gas_exposure)) +
  geom_bar(stat = "identity", alpha = 0.7) +
  scale_fill_manual(values = c("green", "yellow", "red", "grey"), name = "Smoke/Gas Exposure") +
  labs(
    title = "Average BMI by Smoke/Gas Exposure",
    x = "Smoke/Gas Exposure",
    y = "Body Mass Index (BMI)"
  ) +
  theme_minimal() +
  theme(
    legend.position = "none"  # Remove legend if not needed
  )
```

4.  Statistical summary graphs of FEV by BMI and FEV by smoke/gas exposure category.

From the statistical summary graph of FEV by BMI category, we see that there is overlap between the obese and overweight categories which signifies that there is likely to not be difference between the two groups in their association with FEV. However, there is no overlap between normal and underweight categories which signified there there is likely to be a difference or association between these categories and FEV.

```{r}
ggplot(merged_data, aes(x = obesity_level, y = fev, fill = obesity_level)) +
  geom_boxplot() +
  scale_fill_manual(values = c("lightyellow", "lightpink", "red", "orange")) +
  labs(title = "Boxplot of FEV by BMI Category",
       x = "BMI Category",
       y = "FEV") +
  theme_minimal()
```

This box plot of FEV by smoke/gas exposure shows overlap between all categories of exposure. Because of this, it is likely that there is no difference between these categories in their association between FEV.

```{r}
ggplot(merged_data, aes(x = smoke_gas_exposure, y = fev, fill = smoke_gas_exposure)) +
  geom_boxplot() +
  scale_fill_manual(values = c("lavender", "lightyellow", "pink", "skyblue")) +
  labs(title = "Boxplot of FEV by Smoke/Gas Exposure",
       x = "Smoke/Gas Exposure",
       y = "FEV") +
  theme_minimal()
```

5.  A leaflet map showing the concentrations of PM2.5 mass in each of the CHS communities. The map shows greater PM 2.5 concentrations in the Mira Loma area with a concentration of 29.97 µg/m3. Lompoc seems to have the lowest PM 2.5 concentration with 5.96 µg/m3.

```{r}
library(leaflet)
leaflet(merged_data) |>
addTiles() |>
addCircles(lng = ~lon, lat = ~lat, weight = 1,
radius = ~pm25_mass * 100, popup = ~paste(townname, "<br>PM2.5:", pm25_mass),
color = "darkblue", fillOpacity = 0.5) |>
addLegend("bottomright", pal = colorNumeric("Blues", merged_data$pm25_mass), 
values = merged_data$pm25_mass, title = "PM2.5 Concentration")
```

6.  Choose a visualization to examine whether PM2.5 mass is associated with FEV. The scatterplot shows that there seems to be a slightly negative association between pm2.5 mass and FEV. We can see that as mass increases, FEV slightly decreases.

```{r}
ggplot(merged_data, aes(x = pm25_mass, y = fev)) +
  geom_point(alpha = 0.5) +
  geom_smooth(method = "lm", se = FALSE, color = "red") +
  labs(title = "Relationship between PM2.5 Mass and FEV",
       x = "PM2.5 Mass",
       y = "FEV") +
  theme_minimal()
```