Aviation Incidents in US
2025-10-30
Project Overview
Based on data from the National Transportation Safety Board (NTSB)
repository, this project provides a structured and data-driven
investigation of aviation accidents across the United States.
The goal is to derive valuable insights into the characteristics,
causes, and consequences of these incidents.
- Applies exploratory data analysis (EDA), hypothesis testing, and feature extraction to examine aviation accident data.
- Focuses on identifying statistically significant patterns that reveal key aviation risk factors.
- Key variables explored include:
- Aircraft type
- Flight purpose
- Weather Condition
- Time of day
- Geographical location
- Aims to understand how these variables influence the frequency and severity of accidents.
- The project not only offers a retrospective view but also lays the groundwork for future predictive modeling.
- Insights are intended to support aviation authorities, operators, and policymakers in bridging safety gaps using evidence-based strategies.
Load the required libraries
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.4 ✔ tidyr 1.3.1
## ✔ purrr 1.0.4
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(dplyr)——————- Data Preprocessing ———————————
# Load the datasets
crash_data <- read_csv("C:/PaNDa/not possible/CAP_482/Project_DataSets/aviation.csv")## Rows: 44507 Columns: 36
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (26): NtsbNo, EventType, City, State, Country, ReportNo, N, ReportType,...
## dbl (6): Mkey, FatalInjuryCount, SeriousInjuryCount, MinorInjuryCount, Lat...
## lgl (2): HasSafetyRec, RepGenFlag
## dttm (2): EventDate, OriginalPublishDate
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.View(crash_data)# Missing values
colSums(is.na(crash_data))## NtsbNo EventType Mkey EventDate
## 0 0 0 0
## City State Country ReportNo
## 28 6536 43 44413
## N HasSafetyRec ReportType OriginalPublishDate
## 87 0 0 6176
## HighestInjuryLevel FatalInjuryCount SeriousInjuryCount MinorInjuryCount
## 746 0 0 0
## ProbableCause Latitude Longitude Make
## 7131 0 0 52
## Model AirCraftCategory AirportID AirportName
## 65 441 17243 17149
## AmateurBuilt NumberOfEngines Scheduled PurposeOfFlight
## 0 5275 39231 6801
## FAR AirCraftDamage WeatherCondition Operator
## 644 338 5130 24135
## ReportStatus RepGenFlag DocketUrl DocketPublishDate
## 0 0 21116 21116# Remove columns with more than 50% missing values
crash_data <- crash_data %>%
select(-c(DocketUrl, DocketPublishDate))
View(crash_data)# Fill Missing Numeric Values with Mean/Median
crash_data <- crash_data %>%
mutate(across(where(is.numeric), ~ ifelse(is.na(.), mean(., na.rm = TRUE), .)))
View(crash_data)# Fill Missing Categorical Values with Mode
crash_data <- crash_data %>%
mutate(across(where(is.character), ~ ifelse(is.na(.), names(sort(table(.), decreasing = TRUE))[1], .)))
View(crash_data)# Convert categorical variables to factors
crash_data <- crash_data %>%
mutate(across(where(is.character), as.factor))
str(crash_data)## tibble [44,507 × 34] (S3: tbl_df/tbl/data.frame)
## $ NtsbNo : Factor w/ 44507 levels "ANC00FA024","ANC00FA052",..: 25234 2575 44506 11169 25169 15984 16017 16018 11168 11166 ...
## $ EventType : Factor w/ 3 levels "ACC","INC","OCC": 1 1 1 1 1 1 1 1 1 1 ...
## $ Mkey : num [1:44507] 199500 199498 199524 199496 199492 ...
## $ EventDate : POSIXct[1:44507], format: "2025-01-01 02:20:00" "2024-12-31 14:30:00" ...
## $ City : Factor w/ 13793 levels "40 nm vicinity south of Lake Jackson",..: 8366 307 7010 4333 9397 4842 8222 9088 8017 9317 ...
## $ State : Factor w/ 57 levels "Alabama","Alaska",..: 12 2 35 50 39 6 6 6 19 20 ...
## $ Country : Factor w/ 190 levels "Afghanistan",..: 179 179 179 179 179 179 141 128 179 179 ...
## $ ReportNo : Factor w/ 94 levels "AAB0202","AAB0203",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ N : Factor w/ 42077 levels "(H-VISTA",",",..: 33019 14854 28743 18403 28076 26450 2592 39899 19803 29142 ...
## $ HasSafetyRec : logi [1:44507] FALSE FALSE FALSE FALSE FALSE FALSE ...
## $ ReportType : Factor w/ 4 levels "BoardBrief","DirectorBrief",..: 2 2 2 2 2 2 3 2 2 2 ...
## $ OriginalPublishDate: POSIXct[1:44507], format: NA NA ...
## $ HighestInjuryLevel : Factor w/ 4 levels "Fatal","Minor",..: 3 3 3 3 1 4 1 3 3 4 ...
## $ FatalInjuryCount : num [1:44507] 0 0 0 0 1 0 173 0 0 0 ...
## $ SeriousInjuryCount : num [1:44507] 0 0 0 0 0 1 2 0 0 1 ...
## $ MinorInjuryCount : num [1:44507] 0 0 0 0 0 0 0 0 0 0 ...
## $ ProbableCause : Factor w/ 34533 levels "'THIS CASE WAS MODIFIED MAY 30, 2006.'The airplane's inadvertent impact with one of several deer that had enter"| __truncated__,..: 1443 1443 1443 29635 1443 1443 1443 1443 1443 1443 ...
## $ Latitude : num [1:44507] 26.2 61.2 35.9 29.3 39 ...
## $ Longitude : num [1:44507] -81.8 -149.8 -106.3 -94.8 -83.4 ...
## $ Make : Factor w/ 6463 levels ",","107.5 Flying Corporation",..: 652 1227 1227 4963 1227 812 812 812 1826 1553 ...
## $ Model : Factor w/ 8860 levels "-","-269C","(EX) RV-6",..: 1739 260 365 6730 664 1354 1093 1093 3405 3340 ...
## $ AirCraftCategory : Factor w/ 38 levels ",",",AIR,AIR",..: 3 3 3 21 3 3 3 3 3 3 ...
## $ AirportID : Factor w/ 7691 levels "-","---","(AZ38)",..: 2154 2132 4953 5780 5781 5781 5781 5781 5622 5781 ...
## $ AirportName : Factor w/ 14914 levels "---","-70.8301542",..: 9243 8574 10775 10815 10775 10775 10775 10775 10775 10775 ...
## $ AmateurBuilt : Factor w/ 7 levels "FALSE","FALSE,FALSE",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ NumberOfEngines : Factor w/ 34 levels ",",",,",",0",..: 11 11 11 11 18 11 11 11 11 11 ...
## $ Scheduled : Factor w/ 3 levels "NSCH","SCHD",..: 1 1 1 1 1 2 2 1 1 1 ...
## $ PurposeOfFlight : Factor w/ 99 levels ",",",,",",AOBV",..: 61 48 61 75 37 61 61 61 61 61 ...
## $ FAR : Factor w/ 69 levels ",","091,ARMF",..: 39 39 39 39 39 26 48 39 39 39 ...
## $ AirCraftDamage : Factor w/ 28 levels ",","Destroyed",..: 19 19 19 19 19 19 19 19 19 19 ...
## $ WeatherCondition : Factor w/ 3 levels "IMC","Unknown",..: 1 3 3 3 3 3 3 3 3 3 ...
## $ Operator : Factor w/ 16956 levels "--","– M/s Jindal Steel & Power Ltd.",..: 11965 11965 11965 6228 11965 15729 8175 9086 9736 11965 ...
## $ ReportStatus : Factor w/ 3 levels "Completed","In work",..: 2 2 2 1 2 2 3 3 2 2 ...
## $ RepGenFlag : logi [1:44507] FALSE FALSE FALSE FALSE FALSE FALSE ...# Convert Manufacturer in uppercase
crash_data <- crash_data %>%
mutate(Make = toupper(Make))# Replace empty/blank category "," with "UNKNOWN" in AirCraftCategory
crash_data <- crash_data %>%
mutate(AirCraftCategory = ifelse(AirCraftCategory == " , ", "UNKNOWN", AirCraftCategory))# Save the cleaned datasets
write_csv(crash_data, "C:/PaNDa/not possible/CAP_482/Project_DataSets/aviation_cleaned.csv")——————- Exploratory Data Analysis ——————
Q1. What is the average number of fatalities per incident by aircraft category
avg_fatalities_by_category <- crash_data %>%
group_by(AirCraftCategory) %>%
summarise(AvgFatalities = mean(FatalInjuryCount, na.rm = TRUE)) %>%
arrange(desc(AvgFatalities))
View(avg_fatalities_by_category)Interpretation
measures the severity of accidents for various aircraft types.
aids in the identification of high-risk groups for targeted safety measures.
Q2. What percentage of incidents have an official report published?
report_percentage <- crash_data %>%
summarise(ReportPublished = sum(!is.na(ReportStatus)) / n() * 100)
cat("Percentage of incidents with an official report published: ", report_percentage$ReportPublished, "%\n")## Percentage of incidents with an official report published: 100 %Interpretation
shows the fullness of the dataset and the transparency of the regulations.
Low rates could be an indication of incomplete investigations or underreporting
Q3. What are the most common causes of aviation accidents?
commonn_causes <- crash_data %>%
group_by(ProbableCause) %>%
summarise(Count = n()) %>%
arrange(desc(Count)) %>%
top_n(10, Count)
View(commonn_causes)Interpretation
identifies the leading causes of aviation accidents.
This information can be used to target safety measures and regulations for specific causes.
——————- Data Extraction and Filtering ——————
Q4. Which years had the highest aviation accident rates?
accident_rates_by_year <- crash_data %>%
mutate(Year = as.numeric(substr(EventDate, 1, 4))) %>%
count(Year) %>%
arrange(desc(n))
View(accident_rates_by_year)Interpretation
The years with the highest accident rates indicate periods of increased risk in aviation safety.
This information can be used to analyze trends, identify potential causes, and implement safety measures.
Q5. What is the survival rate of aviation incidents?
survival_rate <- crash_data %>%
mutate(Survival = (1 - (FatalInjuryCount / (FatalInjuryCount + SeriousInjuryCount + MinorInjuryCount ))) * 100)
cat("Survival rate of aviation incidents: ", mean(survival_rate$Survival, na.rm = TRUE), "%\n")## Survival rate of aviation incidents: 58.05775 %Interpretation
The survival rate of aviation incidents indicates the percentage of survivors in accidents.
This information can be used to assess the effectiveness of safety measures and regulations.
Q6. Do weather conditions (VMC vs IMC) contribute to more accidents?
weather_accidents <- crash_data %>%
group_by(WeatherCondition) %>%
summarise(Count = n()) %>%
arrange(desc(Count))
cat("Accidents in VMC: ", weather_accidents$Count[weather_accidents$WeatherCondition == "VMC"], "\n")## Accidents in VMC: 41905cat("Accidents in IMC: ", weather_accidents$Count[weather_accidents$WeatherCondition == "IMC"], "\n")## Accidents in IMC: 2254cat("Accidents in Unknown: ", weather_accidents$Count[weather_accidents$WeatherCondition == "Unknown"], "\n")## Accidents in Unknown: 348Interpretation
The number of accidents in VMC and IMC conditions indicates the impact of weather on aviation safety.
This information can be used to target safety measures and regulations for specific weather conditions.
Q7. How many incidents involve multi-engine aircraft?
multi_engine_incidents <- crash_data %>%
count(NumberOfEngines) %>%
summarise(TotalIncidents = sum(n))
cat("Total incidents involving multi-engine aircraft: ", multi_engine_incidents$TotalIncidents, "\n")## Total incidents involving multi-engine aircraft: 44507Interpretation
The number of incidents involving multi-engine aircraft indicates the prevalence of these aircraft in aviation accidents.
This information can be used to assess the safety of multi-engine aircraft and target safety measures accordingly.
——————- Grouping and Summarization ——————
Q8. Which manufacturer has the highest number of fatal incidents per 100 aircraft registered?
fatal_incidents_per_manufacturer <- crash_data %>%
group_by(Make) %>%
summarise(Fatal_Incidents = sum(FatalInjuryCount, na.rm = TRUE)) %>%
arrange(desc(Fatal_Incidents)) %>%
head(10)
cat("Top 10 manufacturers with the highest number of fatal incidents:\n")## Top 10 manufacturers with the highest number of fatal incidents:print(fatal_incidents_per_manufacturer)## # A tibble: 10 × 2
## Make Fatal_Incidents
## <chr> <dbl>
## 1 BOEING 4950
## 2 CESSNA 4253
## 3 PIPER 2759
## 4 BEECH 1802
## 5 AIRBUS 1330
## 6 AIRBUS INDUSTRIE 1088
## 7 BELL 738
## 8 ROBINSON 548
## 9 MOONEY 285
## 10 MCDONNELL DOUGLAS 264Interpretation
The top ten manufacturers with the highest number of fatal incidents indicate which manufacturers are more prone to accidents.
This information can be used to target safety measures and regulations for specific manufacturers.
Q9. Which type of flight purpose has the highest accident rate per 1000 flights?
accident_rate_by_purpose <- crash_data %>%
group_by(PurposeOfFlight) %>%
summarise(Total_Incidents = n()) %>%
arrange(desc(Total_Incidents))
cat("Accident rate by purpose of flight:\n")## Accident rate by purpose of flight:print(accident_rate_by_purpose)## # A tibble: 99 × 2
## PurposeOfFlight Total_Incidents
## <fct> <int>
## 1 PERS 30667
## 2 INST 5176
## 3 AAPL 1869
## 4 BUS 1159
## 5 POSI 955
## 6 UNK 821
## 7 OWRK 680
## 8 FLTS 462
## 9 AOBV 459
## 10 PUBU 238
## # ℹ 89 more rowsInterpretation
The accident rate by purpose of flight indicates which types of flights are more prone to accidents.
This information can be used to target safety measures and regulations for specific flight purposes.
Q10. Which type of aircraft is most frequently involved in fatal incidents?
fatal_incidents_by_aircraft <- crash_data %>%
group_by(AirCraftCategory) %>%
summarise(Fatal_Incidents = sum(FatalInjuryCount, na.rm = TRUE)) %>%
arrange(desc(Fatal_Incidents))
cat("Top aircraft categories involved in fatal incidents:\n")## Top aircraft categories involved in fatal incidents:print(fatal_incidents_by_aircraft)## # A tibble: 38 × 2
## AirCraftCategory Fatal_Incidents
## <int> <dbl>
## 1 3 23613
## 2 21 2315
## 3 5 502
## 4 1 191
## 5 17 134
## 6 38 84
## 7 20 80
## 8 12 55
## 9 9 42
## 10 23 24
## # ℹ 28 more rowsInterpretation
The top aircraft categories involved in fatal incidents indicate the types of aircraft that are more prone to accidents.
This information can be used to target safety measures and regulations for specific aircraft types.
——————- Sorting and Ranking ——————
Q11. Which state has the highest number of aviation accidents per million people?
accidents_per_state <- crash_data %>%
group_by(State) %>%
summarise(Total_Incidents = sum(FatalInjuryCount, na.rm = TRUE)) %>%
arrange(desc(Total_Incidents)) %>%
head(10)
cat("Top 10 states with the highest number of aviation accidents:\n")## Top 10 states with the highest number of aviation accidents:print(accidents_per_state)## # A tibble: 10 × 2
## State Total_Incidents
## <fct> <dbl>
## 1 California 16400
## 2 Florida 860
## 3 Texas 836
## 4 New York 713
## 5 Alaska 507
## 6 Arizona 443
## 7 Colorado 418
## 8 Georgia 388
## 9 North Carolina 296
## 10 Utah 272Interpretation
The top ten states with the highest number of aviation accidents indicate regions with increased aviation activity or risk.
This information can be used to target safety campaigns or regulations in those areas.
Q12. Rank the top 5 airline with the least accidents
least_accidents <- crash_data %>%
group_by(Make) %>%
summarise(Total_Incidents = n()) %>%
arrange(Total_Incidents) %>%
head(5)
cat("Top 5 airlines with the least accidents:\n")## Top 5 airlines with the least accidents:print(least_accidents)## # A tibble: 5 × 2
## Make Total_Incidents
## <chr> <int>
## 1 , 1
## 2 107.5 FLYING CORPORATION 1
## 3 1200 1
## 4 177MF LLC 1
## 5 1977 COLFER-CHAN 1Interpretation
The top five airlines with the least number of accidents indicate a strong safety record.
This information can be used to identify best practices and safety measures that can be adopted by other airlines.
Q13. Which five years had the deadliest aviation accidents?
deadliest_years <- crash_data %>%
mutate(Year = as.numeric(substr(EventDate, 1, 4))) %>%
group_by(Year) %>%
summarise(Total_Fatalities = sum(FatalInjuryCount, na.rm = TRUE)) %>%
arrange(desc(Total_Fatalities)) %>%
head(5)
cat("Top 5 deadliest years for aviation accidents:\n")## Top 5 deadliest years for aviation accidents:print(deadliest_years)## # A tibble: 5 × 2
## Year Total_Fatalities
## <dbl> <dbl>
## 1 2000 1716
## 2 2005 1674
## 3 2001 1564
## 4 2010 1374
## 5 2003 1347View(crash_data)Interpretation
The top five years with the highest number of fatalities indicate periods of increased risk in aviation safety.
This information can be used to analyze trends, identify potential causes, and implement safety measures.
——————- Feature Engineering ——————
Q14. Create a new column for “IncidentSeverity” based on the number of fatalities
crash_data <- crash_data %>%
mutate(IncidentSeverity = case_when(
FatalInjuryCount > 0 ~ "Catastrophic",
SeriousInjuryCount > 0 ~ "Serious",
MinorInjuryCount > 0 ~ "Minor",
TRUE ~ "No Injury"
))
View(crash_data)Interpretation
divides events into four categories: minor, serious, catastrophic, and injury-free.
promotes risk segmentation and improves readability.
Q15. Generate a new feature ‘FatalityRate’ as the ratio of fatalities to total injuries
crash_data <- crash_data %>%
mutate(FatalityRate = (FatalInjuryCount / (FatalInjuryCount + SeriousInjuryCount + MinorInjuryCount)) * 100)
View(crash_data)Interpretation:
The new feature ‘FatalityRate’ provides a percentage representation of fatalities relative to total injuries.
This metric can be used to assess the severity of incidents and compare across different categories or conditions.
——————- Hypothesis Testing and Advanced Insights ——————
Q16. Does the time of year affect the number of aviation accidents?
crash_data$EventMonth <- as.numeric(format(as.Date(crash_data$EventDate, format="%Y-%m-%d"), "%m"))
accidents_by_month <- crash_data %>%
group_by(EventMonth) %>%
summarise(Total_Incidents = n()) %>%
arrange(EventMonth)
cat("Accidents by month:\n")## Accidents by month:print(accidents_by_month)## # A tibble: 12 × 2
## EventMonth Total_Incidents
## <dbl> <int>
## 1 1 2519
## 2 2 2638
## 3 3 3290
## 4 4 3536
## 5 5 4221
## 6 6 4861
## 7 7 5384
## 8 8 4890
## 9 9 4189
## 10 10 3616
## 11 11 2779
## 12 12 2584Interpretation:
The bar chart shows the number of incidents per month.
Peaks in certain months may indicate seasonal trends, such as increased flying during summer or holiday seasons.
Q17. Are amateur-built aircraft more dangerous than factory-built aircraft?
# Step 1: Keep only rows with valid TRUE/FALSE values
crash_data$AmateurBuilt <- ifelse(crash_data$AmateurBuilt == "TRUE", TRUE,
ifelse(crash_data$AmateurBuilt == "FALSE", FALSE, NA))
aviation_clean <- crash_data %>%
filter(!is.na(AmateurBuilt))
t.test(FatalInjuryCount ~ AmateurBuilt, data = crash_data)##
## Welch Two Sample t-test
##
## data: FatalInjuryCount by AmateurBuilt
## t = 10.462, df = 43580, p-value < 2.2e-16
## alternative hypothesis: true difference in means between group FALSE and group TRUE is not equal to 0
## 95 percent confidence interval:
## 0.2513898 0.3672959
## sample estimates:
## mean in group FALSE mean in group TRUE
## 0.6364457 0.3271028Interpretation:
The t-test compares the average number of fatalities between amateur-built and factory-built aircraft.
A significant p-value (typically < 0.05) indicates a meaningful difference in fatality rates between the two types of aircraft.
Q18. Is there a significant difference in fatal injuries between incidents that occurred in the Northern Hemisphere vs. the Southern Hemisphere?
# Step 1: Classify hemisphere
aviation_geo <- crash_data %>%
filter(!is.na(Latitude), !is.na(FatalInjuryCount)) %>%
mutate(Hemisphere = ifelse(Latitude >= 0, "Northern", "Southern"))# Step 2: Perform t-test
t.test(FatalInjuryCount ~ Hemisphere, data = aviation_geo)##
## Welch Two Sample t-test
##
## data: FatalInjuryCount by Hemisphere
## t = -6.1537, df = 899.67, p-value = 1.139e-09
## alternative hypothesis: true difference in means between group Northern and group Southern is not equal to 0
## 95 percent confidence interval:
## -1.6733180 -0.8640648
## sample estimates:
## mean in group Northern mean in group Southern
## 0.5848555 1.8535469Interpretation:
The t-test compares the average number of fatalities between the Northern and Southern Hemispheres.
A significant p-value (typically < 0.05) indicates a meaningful difference in fatality rates between the two hemispheres.
Q19. Do incidents at airports have a higher fatality rate than those that occur elsewhere?
aviation_airport <- crash_data %>%
filter(!is.na(AirportID)) %>%
group_by(AirportID) %>%
summarise(Total_Fatalities = mean(FatalInjuryCount, na.rm = TRUE)) %>%
arrange(desc(Total_Fatalities))
cat("Fatality rate at airports:\n")## Fatality rate at airports:print(aviation_airport)## # A tibble: 7,691 × 2
## AirportID Total_Fatalities
## <fct> <dbl>
## 1 OPRN 157
## 2 FMCH 152
## 3 URSS 113
## 4 MUHA 112
## 5 OLBA 90
## 6 XUBS 89
## 7 CGK 62
## 8 RCQC 58
## 9 WIHH 44
## 10 ZYLD 42
## # ℹ 7,681 more rowsInterpretation:
evaluates the severity of collisions near and far from airports.
can emphasize how crucial it is to be close to emergency assistance.
——————- Visualization ——————
Q20. How have aviation incidents changed over the years.
ggplot(accident_rates_by_year, aes(x = Year, y = n)) +
geom_line(color = "steelblue", size = 1) +
geom_point(color = "darkred", size = 2) +
labs(title = "Trend of Aviation Incidents Over the Years",
x = "Year", y = "Incident Count") +
theme_minimal()## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
View(crash_data)Interpretation:
Reveals long-term patterns in incident frequency.
An upward or downward trend can reflect improvements in aviation technology, regulations, or reporting practices.
Spikes may correspond to systemic failures or external events.
Q21. What is the distribution of FatalityRate across AirCraftCategory?
ggplot(crash_data, aes(x = FatalityRate, fill = AirCraftCategory)) +
geom_histogram(binwidth = 5, position = "identity", alpha = 0.7) +
labs(title = "Distribution of Fatality Rate by Aircraft Category",
x = "Fatality Rate (%)", y = "Count") +
theme_minimal() +
scale_fill_brewer(palette = "Set3")## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_bin()`).## Warning: The following aesthetics were dropped during statistical transformation: fill.
## ℹ This can happen when ggplot fails to infer the correct grouping structure in
## the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
## variable into a factor?
Interpretation:
The histogram shows how fatality rates vary across different aircraft categories.
Peaks indicate categories with higher fatality rates, while valleys suggest safer categories.
Q22. How does the number of incidents vary by state?
ggplot(accidents_per_state, aes(x = reorder(State, -Total_Incidents), y = Total_Incidents)) +
geom_bar(stat = "identity", fill = "steelblue", alpha = 0.7) +
coord_flip() + # Rotate for better readability
labs(title = "Number of Aviation Incidents by State",
x = "State", y = "Incident Count") +
theme_minimal()
Interpretation:
The bar chart illustrates the distribution of aviation incidents across states.
States with higher bars indicate more incidents, which may correlate with population density, air traffic, or geographical factors.
Q23. What are the monthly patterns of aviation incidents?
crash_data$MonthName <- month.name[as.integer(crash_data$EventMonth)]
ggplot(crash_data, aes(x = factor(MonthName, levels = month.name))) +
geom_bar(fill = "purple") +
labs(title = "Monthly Distribution of Incidents", x = "Month", y = "Count") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
Interpretation:
The bar chart shows the number of incidents per month.
Peaks in certain months may indicate seasonal trends, such as increased flying during summer or holiday seasons.
Q24. Compare average FatalityRate by WeatherCondition.
avg_fatality_weather <- crash_data %>%
group_by(WeatherCondition) %>%
summarise(AvgFatalityRate = mean(FatalityRate, na.rm = TRUE))
ggplot(avg_fatality_weather, aes(x = WeatherCondition, y = AvgFatalityRate, fill = WeatherCondition)) +
geom_bar(stat = "identity") +
labs(title = "Average Fatality Rate by Weather Condition", x = "Weather Condition", y = "Fatality Rate (%)") +
theme_minimal()
Interpretation:
The bar chart compares average fatality rates across different weather conditions.
Higher bars indicate conditions associated with more severe incidents, while lower bars suggest safer flying conditions.
——————- Regression-Based Questions ——————
Q25. Is there a relationship between NumberOfEngines and FatalityRate?
model1 <- lm(FatalityRate ~ NumberOfEngines, data = crash_data)
summary(model1)##
## Call:
## lm(formula = FatalityRate ~ NumberOfEngines, data = crash_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -83.33 -41.33 -41.33 58.67 82.57
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 83.079 8.010 10.372 < 2e-16 ***
## NumberOfEngines,0 -83.079 48.061 -1.729 0.0839 .
## NumberOfEngines,1 -21.968 17.711 -1.240 0.2149
## NumberOfEngines,2 -83.079 48.061 -1.729 0.0839 .
## NumberOfEngines0 -65.649 8.393 -7.822 5.44e-15 ***
## NumberOfEngines0,0 -33.079 34.453 -0.960 0.3370
## NumberOfEngines0,1 -33.079 34.453 -0.960 0.3370
## NumberOfEngines1 -41.746 8.018 -5.206 1.95e-07 ***
## NumberOfEngines1, -26.829 18.571 -1.445 0.1486
## NumberOfEngines1,0 -33.079 48.061 -0.688 0.4913
## NumberOfEngines1,1 -23.219 8.902 -2.608 0.0091 **
## NumberOfEngines1,1,0 16.921 48.061 0.352 0.7248
## NumberOfEngines1,2 0.254 17.711 0.014 0.9886
## NumberOfEngines2 -34.743 8.071 -4.305 1.68e-05 ***
## NumberOfEngines2, 16.921 34.453 0.491 0.6233
## NumberOfEngines2,0 -83.079 48.061 -1.729 0.0839 .
## NumberOfEngines2,1 3.853 18.571 0.207 0.8357
## NumberOfEngines2,2 -37.246 20.939 -1.779 0.0753 .
## NumberOfEngines2,3 -83.079 48.061 -1.729 0.0839 .
## NumberOfEngines3 -58.377 12.559 -4.648 3.37e-06 ***
## NumberOfEngines4 -48.464 11.034 -4.392 1.13e-05 ***
## NumberOfEngines4,1 16.921 48.061 0.352 0.7248
## NumberOfEngines8 -83.079 34.453 -2.411 0.0159 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 47.39 on 20196 degrees of freedom
## (24288 observations deleted due to missingness)
## Multiple R-squared: 0.01076, Adjusted R-squared: 0.009684
## F-statistic: 9.986 on 22 and 20196 DF, p-value: < 2.2e-16# Visualize Relationship Between NumberOfEngines and FatalityRate
ggplot(crash_data, aes(x = NumberOfEngines, y = FatalityRate)) +
geom_point(alpha = 0.5) +
geom_smooth(method = "lm", col = "red") +
labs(title = "Number of Engines vs. Fatality Rate",
x = "Number of Engines", y = "Fatality Rate (%)") +
theme_minimal()## `geom_smooth()` using formula = 'y ~ x'## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_smooth()`).## Warning: Removed 24288 rows containing missing values or values outside the scale range
## (`geom_point()`).
Interpretation:
The linear regression model indicates the relationship between the number of engines and the fatality rate.
A positive slope suggests that more engines are associated with higher fatality rates, while a negative slope indicates the opposite.
Q26. Can we predict FatalityRate using WeatherCondition, PurposeOfFlight, and NumberOfEngines?
model3 <- lm(FatalityRate ~ WeatherCondition + PurposeOfFlight + NumberOfEngines, data = crash_data)
summary(model3)##
## Call:
## lm(formula = FatalityRate ~ WeatherCondition + PurposeOfFlight +
## NumberOfEngines, data = crash_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -94.98 -39.45 -28.60 60.55 92.52
##
## Coefficients: (5 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.207e+02 1.050e+01 11.502 < 2e-16 ***
## WeatherConditionUnknown -5.647e+00 3.259e+00 -1.733 0.08318 .
## WeatherConditionVMC -3.323e+01 1.178e+00 -28.205 < 2e-16 ***
## PurposeOfFlight,BUS -8.751e+01 4.719e+01 -1.854 0.06368 .
## PurposeOfFlight,FERY 5.247e-01 3.660e+01 0.014 0.98856
## PurposeOfFlight,INST -8.328e+01 3.395e+01 -2.453 0.01418 *
## PurposeOfFlight,OWRK 2.095e+01 3.461e+01 0.605 0.54497
## PurposeOfFlight,PERS -2.031e+01 2.676e+01 -0.759 0.44787
## PurposeOfFlightAAPL -9.769e+01 3.427e+01 -2.850 0.00437 **
## PurposeOfFlightAAPL, -2.905e+01 4.751e+01 -0.611 0.54091
## PurposeOfFlightAAPL,AAPL -2.559e+01 1.743e+01 -1.468 0.14202
## PurposeOfFlightAAPL,PERS 1.249e+01 4.719e+01 0.265 0.79135
## PurposeOfFlightADRP -8.821e+01 3.905e+01 -2.259 0.02390 *
## PurposeOfFlightADRP, -3.732e+01 4.729e+01 -0.789 0.43005
## PurposeOfFlightAOBV -9.118e+01 3.434e+01 -2.655 0.00793 **
## PurposeOfFlightAOBV,AOBV -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightASHO -7.407e+01 3.452e+01 -2.145 0.03193 *
## PurposeOfFlightASHO,ASHO -1.022e+01 1.752e+01 -0.583 0.55976
## PurposeOfFlightBANT -1.004e+02 3.474e+01 -2.889 0.00386 **
## PurposeOfFlightBUS -8.751e+01 3.417e+01 -2.561 0.01045 *
## PurposeOfFlightBUS,BUS -4.572e+01 3.461e+01 -1.321 0.18657
## PurposeOfFlightBUS,INST -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightEXEC -8.844e+01 3.457e+01 -2.558 0.01053 *
## PurposeOfFlightEXEC, 1.249e+01 4.719e+01 0.265 0.79135
## PurposeOfFlightEXLD -9.649e+01 3.462e+01 -2.787 0.00533 **
## PurposeOfFlightFERY -8.540e+01 3.453e+01 -2.473 0.01339 *
## PurposeOfFlightFERY,PERS 2.095e+01 4.751e+01 0.441 0.65922
## PurposeOfFlightFIRF -7.713e+01 3.508e+01 -2.198 0.02793 *
## PurposeOfFlightFIRF,FIRF 2.095e+01 4.751e+01 0.441 0.65922
## PurposeOfFlightFLTS -9.083e+01 3.436e+01 -2.643 0.00822 **
## PurposeOfFlightGLDT -8.479e+01 3.522e+01 -2.407 0.01608 *
## PurposeOfFlightGLDT,PERS,GLDT 1.249e+01 4.719e+01 0.265 0.79135
## PurposeOfFlightINST -1.013e+02 3.425e+01 -2.957 0.00311 **
## PurposeOfFlightINST, 2.095e+01 4.751e+01 0.441 0.65922
## PurposeOfFlightINST,INST 4.584e+00 1.585e+01 0.289 0.77240
## PurposeOfFlightINST,OWRK -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightINST,PERS -1.379e+01 1.502e+01 -0.919 0.35834
## PurposeOfFlightOWRK -9.911e+01 3.431e+01 -2.888 0.00388 **
## PurposeOfFlightOWRK,INST -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightOWRK,OWRK 2.864e+01 2.859e+01 1.002 0.31645
## PurposeOfFlightOWRK,PERS -3.751e+01 4.719e+01 -0.795 0.42665
## PurposeOfFlightPERS -9.043e+01 3.423e+01 -2.642 0.00826 **
## PurposeOfFlightPERS, 1.050e+01 2.527e+01 0.416 0.67769
## PurposeOfFlightPERS,AOBV 2.095e+01 3.461e+01 0.605 0.54497
## PurposeOfFlightPERS,ASHO 2.095e+01 4.751e+01 0.441 0.65922
## PurposeOfFlightPERS,BUS 3.307e+01 5.000e+01 0.661 0.50839
## PurposeOfFlightPERS,GLDT -3.751e+01 3.417e+01 -1.098 0.27233
## PurposeOfFlightPERS,INST 1.282e+01 1.811e+01 0.708 0.47917
## PurposeOfFlightPERS,OWRK 8.529e+00 5.093e+01 0.167 0.86701
## PurposeOfFlightPERS,PERS -3.283e+01 1.258e+01 -2.610 0.00906 **
## PurposeOfFlightPERS,POSI -1.026e+01 3.577e+01 -0.287 0.77423
## PurposeOfFlightPERS,UNK -8.282e+00 3.395e+01 -0.244 0.80730
## PurposeOfFlightPOSI -8.809e+01 3.430e+01 -2.568 0.01024 *
## PurposeOfFlightPOSI,BUS -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightPOSI,PERS 6.800e+00 5.077e+01 0.134 0.89345
## PurposeOfFlightPOSI,POSI 2.095e+01 4.751e+01 0.441 0.65922
## PurposeOfFlightPUBF -9.588e+01 3.485e+01 -2.751 0.00594 **
## PurposeOfFlightPUBL -1.127e+02 3.506e+01 -3.214 0.00131 **
## PurposeOfFlightPUBL,PUBL -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightPUBS -1.005e+02 3.516e+01 -2.860 0.00424 **
## PurposeOfFlightPUBS,PERS -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightPUBS,PUBS 8.529e+00 5.093e+01 0.167 0.86701
## PurposeOfFlightPUBU -9.562e+01 3.450e+01 -2.772 0.00558 **
## PurposeOfFlightPUBU,INST 1.249e+01 4.719e+01 0.265 0.79135
## PurposeOfFlightPUBU,PUBU 7.748e+01 5.137e+01 1.508 0.13153
## PurposeOfFlightSKYD -1.000e+02 3.463e+01 -2.888 0.00388 **
## PurposeOfFlightSKYD,SKYD -7.905e+01 4.751e+01 -1.664 0.09615 .
## PurposeOfFlightUNK -6.951e+01 3.429e+01 -2.027 0.04267 *
## PurposeOfFlightUNK,INST -2.417e-11 6.509e+01 0.000 1.00000
## PurposeOfFlightUNK,PERS -2.438e+01 2.878e+01 -0.847 0.39688
## PurposeOfFlightUNK,UNK 1.121e+01 2.168e+01 0.517 0.60528
## NumberOfEngines,0 -7.730e+01 4.879e+01 -1.584 0.11317
## NumberOfEngines,1 -9.417e-01 1.940e+01 -0.049 0.96129
## NumberOfEngines,2 -1.207e+02 4.721e+01 -2.558 0.01054 *
## NumberOfEngines0 2.124e+01 3.270e+01 0.650 0.51588
## NumberOfEngines0,0 -1.599e+01 3.465e+01 -0.461 0.64446
## NumberOfEngines0,1 NA NA NA NA
## NumberOfEngines1 4.236e+01 3.261e+01 1.299 0.19386
## NumberOfEngines1, -2.059e+01 2.001e+01 -1.029 0.30361
## NumberOfEngines1,0 NA NA NA NA
## NumberOfEngines1,1 -8.466e+00 1.096e+01 -0.773 0.43972
## NumberOfEngines1,1,0 NA NA NA NA
## NumberOfEngines1,2 3.956e+00 2.137e+01 0.185 0.85314
## NumberOfEngines2 4.375e+01 3.261e+01 1.341 0.17980
## NumberOfEngines2, 1.249e+01 4.719e+01 0.265 0.79135
## NumberOfEngines2,0 -6.720e+01 5.312e+01 -1.265 0.20586
## NumberOfEngines2,1 5.685e+00 2.097e+01 0.271 0.78631
## NumberOfEngines2,2 -6.499e+01 2.422e+01 -2.683 0.00730 **
## NumberOfEngines2,3 NA NA NA NA
## NumberOfEngines3 2.106e+01 3.393e+01 0.621 0.53484
## NumberOfEngines4 3.398e+01 3.343e+01 1.016 0.30941
## NumberOfEngines4,1 2.270e+01 4.879e+01 0.465 0.64171
## NumberOfEngines8 NA NA NA NA
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 46.02 on 20131 degrees of freedom
## (24288 observations deleted due to missingness)
## Multiple R-squared: 0.06994, Adjusted R-squared: 0.06592
## F-statistic: 17.4 on 87 and 20131 DF, p-value: < 2.2e-16Interpretation:
The multiple regression model assesses the combined effect of weather conditions, flight purpose, and number of engines on the fatality rate.
Significant predictors will have low p-values, indicating a strong relationship with the fatality rate.
Q27. Does latitude (location) influence fatality rate?
model4 <- lm(FatalityRate ~ Latitude, data = crash_data)
summary(model4)##
## Call:
## lm(formula = FatalityRate ~ Latitude, data = crash_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -41.94 -41.94 -41.94 58.06 58.30
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.194e+01 3.349e-01 125.241 <2e-16 ***
## Latitude -9.913e-07 1.151e-06 -0.861 0.389
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 47.62 on 20217 degrees of freedom
## (24288 observations deleted due to missingness)
## Multiple R-squared: 3.668e-05, Adjusted R-squared: -1.278e-05
## F-statistic: 0.7416 on 1 and 20217 DF, p-value: 0.3892Visualize Relationship Between Latitude and FatalityRate
ggplot(crash_data, aes(x = Latitude, y = FatalityRate)) +
geom_point(alpha = 0.5) +
geom_smooth(method = "lm", col = "tomato") +
labs(title = "Latitude vs. Fatality Rate",
x = "Latitude", y = "Fatality Rate (%)") +
theme_minimal()## `geom_smooth()` using formula = 'y ~ x'## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_smooth()`).## Warning: Removed 24288 rows containing missing values or values outside the scale range
## (`geom_point()`).
Interpretation:
The linear regression model examines the relationship between latitude and fatality rate.
A positive slope indicates that higher latitudes are associated with increased fatality rates, while a negative slope suggests the opposite.
Q28. Does the Number of Fatal, Serious, and Minor Injuries Predict the Fatality Rate?
model2 <- lm(FatalityRate ~ FatalInjuryCount + SeriousInjuryCount + MinorInjuryCount, data = crash_data)
summary(model2)##
## Call:
## lm(formula = FatalityRate ~ FatalInjuryCount + SeriousInjuryCount +
## MinorInjuryCount, data = crash_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -326.55 -40.53 -33.35 53.65 573.84
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 43.77816 0.33835 129.39 <2e-16 ***
## FatalInjuryCount 1.28447 0.04137 31.05 <2e-16 ***
## SeriousInjuryCount -5.21321 0.19925 -26.16 <2e-16 ***
## MinorInjuryCount -1.62530 0.08724 -18.63 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 45.41 on 20215 degrees of freedom
## (24288 observations deleted due to missingness)
## Multiple R-squared: 0.09096, Adjusted R-squared: 0.09083
## F-statistic: 674.3 on 3 and 20215 DF, p-value: < 2.2e-16# Visualize Relationship Between FatalInjuryCount and FatalityRate
ggplot(crash_data, aes(x = FatalInjuryCount, y = FatalityRate)) +
geom_point(alpha = 0.5) +
geom_smooth(method = "lm", col = "red") +
labs(title = "Fatal Injury Count vs. Fatality Rate",
x = "Fatal Injury Count", y = "Fatality Rate (%)") +
theme_minimal()## `geom_smooth()` using formula = 'y ~ x'## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_smooth()`).## Warning: Removed 24288 rows containing missing values or values outside the scale range
## (`geom_point()`).
Interpretation:
The multiple regression model evaluates the predictive power of fatal, serious, and minor injuries on the fatality rate.
Significant predictors will have low p-values, indicating a strong relationship with the fatality rate.
Q29. Can Year (from EventDate) explain variation in FatalInjuryCount?
crash_data$Year <- as.numeric(substr(crash_data$EventDate, 1, 4))
model5 <- lm(FatalInjuryCount ~ Year, data = crash_data)
summary(model5)##
## Call:
## lm(formula = FatalInjuryCount ~ Year, data = crash_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.759 -0.667 -0.575 -0.456 297.425
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 27.117413 6.911479 3.924 8.74e-05 ***
## Year -0.013179 0.003436 -3.835 0.000126 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.282 on 44505 degrees of freedom
## Multiple R-squared: 0.0003304, Adjusted R-squared: 0.0003079
## F-statistic: 14.71 on 1 and 44505 DF, p-value: 0.0001256# Visualize Relationship Between Year and FatalInjuryCount
ggplot(crash_data, aes(x = Year, y = FatalInjuryCount)) +
geom_point(alpha = 0.5) +
geom_smooth(method = "lm", col = "blue") +
labs(title = "Year vs. Fatal Injury Count",
x = "Year", y = "Fatal Injury Count") +
theme_minimal()## `geom_smooth()` using formula = 'y ~ x'
str(crash_data)## tibble [44,507 × 39] (S3: tbl_df/tbl/data.frame)
## $ NtsbNo : Factor w/ 44507 levels "ANC00FA024","ANC00FA052",..: 25234 2575 44506 11169 25169 15984 16017 16018 11168 11166 ...
## $ EventType : Factor w/ 3 levels "ACC","INC","OCC": 1 1 1 1 1 1 1 1 1 1 ...
## $ Mkey : num [1:44507] 199500 199498 199524 199496 199492 ...
## $ EventDate : POSIXct[1:44507], format: "2025-01-01 02:20:00" "2024-12-31 14:30:00" ...
## $ City : Factor w/ 13793 levels "40 nm vicinity south of Lake Jackson",..: 8366 307 7010 4333 9397 4842 8222 9088 8017 9317 ...
## $ State : Factor w/ 57 levels "Alabama","Alaska",..: 12 2 35 50 39 6 6 6 19 20 ...
## $ Country : Factor w/ 190 levels "Afghanistan",..: 179 179 179 179 179 179 141 128 179 179 ...
## $ ReportNo : Factor w/ 94 levels "AAB0202","AAB0203",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ N : Factor w/ 42077 levels "(H-VISTA",",",..: 33019 14854 28743 18403 28076 26450 2592 39899 19803 29142 ...
## $ HasSafetyRec : logi [1:44507] FALSE FALSE FALSE FALSE FALSE FALSE ...
## $ ReportType : Factor w/ 4 levels "BoardBrief","DirectorBrief",..: 2 2 2 2 2 2 3 2 2 2 ...
## $ OriginalPublishDate: POSIXct[1:44507], format: NA NA ...
## $ HighestInjuryLevel : Factor w/ 4 levels "Fatal","Minor",..: 3 3 3 3 1 4 1 3 3 4 ...
## $ FatalInjuryCount : num [1:44507] 0 0 0 0 1 0 173 0 0 0 ...
## $ SeriousInjuryCount : num [1:44507] 0 0 0 0 0 1 2 0 0 1 ...
## $ MinorInjuryCount : num [1:44507] 0 0 0 0 0 0 0 0 0 0 ...
## $ ProbableCause : Factor w/ 34533 levels "'THIS CASE WAS MODIFIED MAY 30, 2006.'The airplane's inadvertent impact with one of several deer that had enter"| __truncated__,..: 1443 1443 1443 29635 1443 1443 1443 1443 1443 1443 ...
## $ Latitude : num [1:44507] 26.2 61.2 35.9 29.3 39 ...
## $ Longitude : num [1:44507] -81.8 -149.8 -106.3 -94.8 -83.4 ...
## $ Make : chr [1:44507] "BEECH" "CESSNA" "CESSNA" "ROBINSON HELICOPTER" ...
## $ Model : Factor w/ 8860 levels "-","-269C","(EX) RV-6",..: 1739 260 365 6730 664 1354 1093 1093 3405 3340 ...
## $ AirCraftCategory : int [1:44507] 3 3 3 21 3 3 3 3 3 3 ...
## $ AirportID : Factor w/ 7691 levels "-","---","(AZ38)",..: 2154 2132 4953 5780 5781 5781 5781 5781 5622 5781 ...
## $ AirportName : Factor w/ 14914 levels "---","-70.8301542",..: 9243 8574 10775 10815 10775 10775 10775 10775 10775 10775 ...
## $ AmateurBuilt : logi [1:44507] FALSE FALSE FALSE FALSE FALSE FALSE ...
## $ NumberOfEngines : Factor w/ 34 levels ",",",,",",0",..: 11 11 11 11 18 11 11 11 11 11 ...
## $ Scheduled : Factor w/ 3 levels "NSCH","SCHD",..: 1 1 1 1 1 2 2 1 1 1 ...
## $ PurposeOfFlight : Factor w/ 99 levels ",",",,",",AOBV",..: 61 48 61 75 37 61 61 61 61 61 ...
## $ FAR : Factor w/ 69 levels ",","091,ARMF",..: 39 39 39 39 39 26 48 39 39 39 ...
## $ AirCraftDamage : Factor w/ 28 levels ",","Destroyed",..: 19 19 19 19 19 19 19 19 19 19 ...
## $ WeatherCondition : Factor w/ 3 levels "IMC","Unknown",..: 1 3 3 3 3 3 3 3 3 3 ...
## $ Operator : Factor w/ 16956 levels "--","– M/s Jindal Steel & Power Ltd.",..: 11965 11965 11965 6228 11965 15729 8175 9086 9736 11965 ...
## $ ReportStatus : Factor w/ 3 levels "Completed","In work",..: 2 2 2 1 2 2 3 3 2 2 ...
## $ RepGenFlag : logi [1:44507] FALSE FALSE FALSE FALSE FALSE FALSE ...
## $ IncidentSeverity : chr [1:44507] "No Injury" "No Injury" "No Injury" "No Injury" ...
## $ FatalityRate : num [1:44507] NaN NaN NaN NaN 100 ...
## $ EventMonth : num [1:44507] 1 12 12 12 12 12 12 12 12 12 ...
## $ MonthName : chr [1:44507] "January" "December" "December" "December" ...
## $ Year : num [1:44507] 2025 2024 2024 2024 2024 ...Interpretation:
The linear regression model assesses the relationship between the year of the incident and the number of fatal injuries.
A positive slope indicates that the number of fatal injuries has increased over the years, while a negative slope suggests a decrease.
——————- ANOVA ——————
Q30. Which aircraft category has the highest variation in FatalityRate?
crash_data$AirCraftCategory <- as.factor(crash_data$AirCraftCategory)
anova_model <- aov(FatalityRate ~ AirCraftCategory, data = crash_data)
anova_results <- summary(anova_model)
print(anova_results)## Df Sum Sq Mean Sq F value Pr(>F)
## AirCraftCategory 27 598097 22152 9.884 <2e-16 ***
## Residuals 20191 45250185 2241
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 24288 observations deleted due to missingnessTukeyHSD(anova_model)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = FatalityRate ~ AirCraftCategory, data = crash_data)
##
## $AirCraftCategory
## diff lwr upr p adj
## 3-1 -5.710680e+01 -158.6553507 44.4417594 0.9522062
## 5-1 -3.616783e+01 -138.5360164 66.2003648 0.9999686
## 7-1 -5.244658e-10 -203.0785695 203.0785695 1.0000000
## 8-1 -4.250000e+01 -176.8238478 91.8238478 0.9999966
## 9-1 -1.590909e+01 -130.4608532 98.6426714 1.0000000
## 10-1 -5.245795e-10 -160.5477059 160.5477059 1.0000000
## 12-1 -9.216900e+01 -194.2771381 9.9391353 0.1518235
## 13-1 -1.000000e+02 -260.5477059 60.5477059 0.8658929
## 15-1 -8.000000e+01 -208.4381647 48.4381647 0.8658929
## 16-1 -5.000000e+01 -253.0785695 153.0785695 1.0000000
## 17-1 -6.723485e+01 -169.2059110 34.7362140 0.7850985
## 18-1 -3.333333e+01 -176.9315669 110.2649002 1.0000000
## 19-1 -7.000000e+01 -198.4381647 58.4381647 0.9667616
## 20-1 -5.685654e+01 -159.3552724 45.6421923 0.9591170
## 21-1 -6.042149e+01 -162.0222448 41.1792660 0.9134582
## 22-1 -5.370370e+01 -178.0634219 70.6560145 0.9988106
## 23-1 -2.878788e+01 -143.3396411 85.7638836 1.0000000
## 27-1 -1.000000e+02 -303.0785695 103.0785695 0.9911205
## 28-1 -8.368421e+01 -186.8142859 19.4458648 0.3444616
## 29-1 -5.000000e+01 -253.0785695 153.0785695 1.0000000
## 30-1 -6.666667e+01 -210.2649002 76.9315669 0.9962495
## 31-1 -6.060606e+01 -166.6603846 45.4482634 0.9429121
## 32-1 -5.030500e-10 -203.0785695 203.0785695 1.0000000
## 33-1 -4.204545e+01 -161.1108199 77.0199108 0.9999689
## 34-1 -5.191225e-10 -203.0785695 203.0785695 1.0000000
## 35-1 -1.000000e+02 -228.4381647 28.4381647 0.4365639
## 38-1 -5.894040e+01 -161.4833897 43.6025950 0.9392129
## 5-3 2.093897e+01 7.8659904 34.0119493 0.0000010
## 7-3 5.710680e+01 -118.7697569 232.9833481 0.9999943
## 8-3 1.460680e+01 -73.3395087 102.5531000 1.0000000
## 9-3 4.119770e+01 -11.8472064 94.2426159 0.4422323
## 10-3 5.710680e+01 -67.2604919 181.4740831 0.9968307
## 12-3 -3.506221e+01 -45.9124559 -24.2119557 0.0000000
## 13-3 -4.289320e+01 -167.2604919 81.4740831 0.9999805
## 15-3 -2.289320e+01 -101.5571637 55.7707549 0.9999994
## 16-3 7.106796e+00 -168.7697569 182.9833481 1.0000000
## 17-3 -1.012805e+01 -19.6019082 -0.6541975 0.0199686
## 18-3 2.377346e+01 -77.7750927 125.3220173 1.0000000
## 19-3 -1.289320e+01 -91.5571637 65.7707549 1.0000000
## 20-3 2.502556e-01 -13.8084243 14.3089354 1.0000000
## 21-3 -3.314694e+00 -7.1054511 0.4760636 0.2004908
## 22-3 3.403092e+00 -68.4091345 75.2153183 1.0000000
## 23-3 2.831892e+01 -24.7259943 81.3638280 0.9741422
## 27-3 -4.289320e+01 -218.7697569 132.9833481 1.0000000
## 28-3 -2.657741e+01 -44.6735134 -8.4813163 0.0000181
## 29-3 7.106796e+00 -168.7697569 182.9833481 1.0000000
## 30-3 -9.559871e+00 -111.1084261 91.9886840 1.0000000
## 31-3 -3.499265e+00 -34.1452434 27.1467134 1.0000000
## 32-3 5.710680e+01 -118.7697569 232.9833481 0.9999943
## 33-3 1.506134e+01 -47.1336552 77.2563374 1.0000000
## 34-3 5.710680e+01 -118.7697569 232.9833481 0.9999943
## 35-3 -4.289320e+01 -121.5571637 35.7707549 0.9665705
## 38-3 -1.833602e+00 -16.2114187 12.5442153 1.0000000
## 7-5 3.616783e+01 -140.1832421 212.5188937 1.0000000
## 8-5 -6.332174e+00 -95.2236241 82.5592757 1.0000000
## 9-5 2.025873e+01 -34.3388843 74.8563540 0.9999195
## 10-5 3.616783e+01 -88.8696071 161.2052587 0.9999995
## 12-5 -5.600118e+01 -72.8791300 -39.1232212 0.0000000
## 13-5 -6.383217e+01 -188.8696071 61.2052587 0.9854512
## 15-5 -4.383217e+01 -123.5514059 35.8870575 0.9631086
## 16-5 -1.383217e+01 -190.1832421 162.5188937 1.0000000
## 17-5 -3.106702e+01 -47.0948532 -15.0391922 0.0000000
## 18-5 2.834492e+00 -99.5336981 105.2026831 1.0000000
## 19-5 -3.383217e+01 -113.5514059 45.8870575 0.9991100
## 20-5 -2.068871e+01 -39.7880309 -1.5893977 0.0163409
## 21-5 -2.425366e+01 -37.7261266 -10.7812007 0.0000000
## 22-5 -1.753588e+01 -90.5025366 55.4307808 1.0000000
## 23-5 7.379947e+00 -47.2176721 61.9775662 1.0000000
## 27-5 -6.383217e+01 -240.1832421 112.5188937 0.9999498
## 28-5 -4.751638e+01 -69.7561284 -25.2766410 0.0000000
## 29-5 -1.383217e+01 -190.1832421 162.5188937 1.0000000
## 30-5 -3.049884e+01 -132.8670315 71.8693497 0.9999991
## 31-5 -2.443823e+01 -57.6995247 8.8230551 0.5693730
## 32-5 3.616783e+01 -140.1832421 212.5188937 1.0000000
## 33-5 -5.877629e+00 -69.4020727 57.6468152 1.0000000
## 34-5 3.616783e+01 -140.1832421 212.5188937 1.0000000
## 35-5 -6.383217e+01 -143.5514059 15.8870575 0.3734881
## 38-5 -2.277257e+01 -42.1080063 -3.4371368 0.0038836
## 8-7 -4.250000e+01 -239.1299794 154.1299794 1.0000000
## 9-7 -1.590909e+01 -199.6005685 167.7823867 1.0000000
## 10-7 -1.136868e-13 -215.3973504 215.3973504 1.0000000
## 12-7 -9.216900e+01 -268.3692405 84.0312377 0.9799984
## 13-7 -1.000000e+02 -315.3973504 115.3973504 0.9962495
## 15-7 -8.000000e+01 -272.6572470 112.6572470 0.9993854
## 16-7 -5.000000e+01 -298.7194364 198.7194364 1.0000000
## 17-7 -6.723485e+01 -243.3556885 108.8859915 0.9998630
## 18-7 -3.333333e+01 -236.4119028 169.7452361 1.0000000
## 19-7 -7.000000e+01 -262.6572470 122.6572470 0.9999460
## 20-7 -5.685654e+01 -233.2834168 119.5703366 0.9999951
## 21-7 -6.042149e+01 -236.3281868 115.4852080 0.9999820
## 22-7 -5.370370e+01 -243.6663111 136.2589037 0.9999997
## 23-7 -2.878788e+01 -212.4793564 154.9035988 1.0000000
## 27-7 -1.000000e+02 -348.7194364 148.7194364 0.9996499
## 28-7 -8.368421e+01 -260.4786254 93.1102043 0.9949827
## 29-7 -5.000000e+01 -298.7194364 198.7194364 1.0000000
## 30-7 -6.666667e+01 -269.7452361 136.4119028 0.9999928
## 31-7 -6.060606e+01 -239.1220888 117.9099676 0.9999858
## 32-7 2.141576e-11 -248.7194364 248.7194364 1.0000000
## 33-7 -4.204545e+01 -228.5850319 144.4941228 1.0000000
## 34-7 5.343281e-12 -248.7194364 248.7194364 1.0000000
## 35-7 -1.000000e+02 -292.6572470 92.6572470 0.9818994
## 38-7 -5.894040e+01 -235.3929914 117.5121967 0.9999897
## 9-8 2.659091e+01 -76.0957486 129.2775668 1.0000000
## 10-8 4.250000e+01 -109.8089271 194.8089271 0.9999998
## 12-8 -4.966900e+01 -138.2608464 38.9228436 0.9538196
## 13-8 -5.750000e+01 -209.8089271 94.8089271 0.9998886
## 15-8 -3.750000e+01 -155.4779876 80.4779876 0.9999963
## 16-8 -7.500000e+00 -204.1299794 189.1299794 1.0000000
## 17-8 -2.473485e+01 -113.1686712 63.6989743 0.9999998
## 18-8 9.166667e+00 -125.1571812 143.4905145 1.0000000
## 19-8 -2.750000e+01 -145.4779876 90.4779876 1.0000000
## 20-8 -1.435654e+01 -103.3982918 74.6852116 1.0000000
## 21-8 -1.792149e+01 -105.9280625 70.0850837 1.0000000
## 22-8 -1.120370e+01 -124.7280752 102.3206678 1.0000000
## 23-8 1.371212e+01 -88.9745365 116.3987789 1.0000000
## 27-8 -5.750000e+01 -254.1299794 139.1299794 0.9999994
## 28-8 -4.118421e+01 -130.9519990 48.5835780 0.9968722
## 29-8 -7.500000e+00 -204.1299794 189.1299794 1.0000000
## 30-8 -2.416667e+01 -158.4905145 110.1571812 1.0000000
## 31-8 -1.810606e+01 -111.2186970 75.0065758 1.0000000
## 32-8 4.250000e+01 -154.1299794 239.1299794 1.0000000
## 33-8 4.545455e-01 -107.2441297 108.1532206 1.0000000
## 34-8 4.250000e+01 -154.1299794 239.1299794 1.0000000
## 35-8 -5.750000e+01 -175.4779876 60.4779876 0.9923244
## 38-8 -1.644040e+01 -105.5330945 72.6522998 1.0000000
## 10-9 1.590909e+01 -119.2841760 151.1023579 1.0000000
## 12-9 -7.625991e+01 -130.3683678 -22.1514532 0.0000601
## 13-9 -8.409091e+01 -219.2841760 51.1023579 0.8675462
## 15-9 -6.409091e+01 -158.9487802 30.7669621 0.7428085
## 16-9 -3.409091e+01 -217.7823867 149.6005685 1.0000000
## 17-9 -5.132576e+01 -105.1750952 2.5235800 0.0878026
## 18-9 -1.742424e+01 -131.9760048 97.1275199 1.0000000
## 19-9 -5.409091e+01 -148.9487802 40.7669621 0.9442400
## 20-9 -4.094745e+01 -95.7894377 13.8945393 0.5328199
## 21-9 -4.451240e+01 -97.6571731 8.6323761 0.2789476
## 22-9 -3.779461e+01 -127.0526269 51.4634013 0.9991433
## 23-9 -1.287879e+01 -87.8705196 62.1129438 1.0000000
## 27-9 -8.409091e+01 -267.7823867 99.6005685 0.9969761
## 28-9 -6.777512e+01 -123.7882074 -11.7620318 0.0022929
## 29-9 -3.409091e+01 -217.7823867 149.6005685 1.0000000
## 30-9 -5.075758e+01 -165.3093381 63.7941866 0.9981914
## 31-9 -4.469697e+01 -105.9274622 16.5335228 0.5839572
## 32-9 1.590909e+01 -167.7823867 199.6005685 1.0000000
## 33-9 -2.613636e+01 -107.8567087 55.5839814 0.9999958
## 34-9 1.590909e+01 -167.7823867 199.6005685 1.0000000
## 35-9 -8.409091e+01 -178.9487802 10.7669621 0.1784741
## 38-9 -4.303131e+01 -97.9559716 11.8933587 0.4223383
## 12-10 -9.216900e+01 -216.9936174 32.6556146 0.5581581
## 13-10 -1.000000e+02 -275.8712001 75.8712001 0.9459669
## 15-10 -8.000000e+01 -227.1444030 67.1444030 0.9677157
## 16-10 -5.000000e+01 -265.3973504 165.3973504 1.0000000
## 17-10 -6.723485e+01 -191.9473609 57.4776640 0.9708412
## 18-10 -3.333333e+01 -193.8810392 127.2143725 1.0000000
## 19-10 -7.000000e+01 -217.1444030 77.1444030 0.9945963
## 20-10 -5.685654e+01 -182.0008700 68.2877898 0.9973119
## 21-10 -6.042149e+01 -184.8314033 63.9884244 0.9927015
## 22-10 -5.370370e+01 -197.3019373 89.8945299 0.9999066
## 23-10 -2.878788e+01 -163.9811457 106.4053882 1.0000000
## 27-10 -1.000000e+02 -315.3973504 115.3973504 0.9962495
## 28-10 -8.368421e+01 -209.3461603 41.9777392 0.7683968
## 29-10 -5.000000e+01 -265.3973504 165.3973504 1.0000000
## 30-10 -6.666667e+01 -227.2143725 93.8810392 0.9993854
## 31-10 -6.060606e+01 -188.6788244 67.4667032 0.9950025
## 32-10 2.152944e-11 -215.3973504 215.3973504 1.0000000
## 33-10 -4.204545e+01 -181.0838463 96.9929373 0.9999987
## 34-10 5.456968e-12 -215.3973504 215.3973504 1.0000000
## 35-10 -1.000000e+02 -247.1444030 47.1444030 0.7320684
## 38-10 -5.894040e+01 -184.1209807 66.2401860 0.9953628
## 13-12 -7.830999e+00 -132.6556146 116.9936174 1.0000000
## 15-12 1.216900e+01 -67.2160164 91.5540192 1.0000000
## 16-12 4.216900e+01 -134.0312377 218.3692405 1.0000000
## 17-12 2.493415e+01 10.6612245 39.2070814 0.0000000
## 18-12 5.883567e+01 -43.2724686 160.9438048 0.9375794
## 19-12 2.216900e+01 -57.2160164 101.5540192 0.9999998
## 20-12 3.531246e+01 17.6600102 52.9649125 0.0000000
## 21-12 3.174751e+01 20.4191232 43.0759008 0.0000000
## 22-12 3.846530e+01 -34.1360688 111.0666642 0.9764838
## 23-12 6.338112e+01 9.2726653 117.4895799 0.0043021
## 27-12 -7.830999e+00 -184.0312377 168.3692405 1.0000000
## 28-12 8.484791e+00 -12.5254765 29.4950583 0.9996215
## 29-12 4.216900e+01 -134.0312377 218.3692405 1.0000000
## 30-12 2.550233e+01 -76.6058020 127.6104715 1.0000000
## 31-12 3.156294e+01 -0.8891553 64.0150369 0.0699221
## 32-12 9.216900e+01 -84.0312377 268.3692405 0.9799984
## 33-12 5.012355e+01 -12.9809706 113.2280643 0.3913691
## 34-12 9.216900e+01 -84.0312377 268.3692405 0.9799984
## 35-12 -7.830999e+00 -87.2160164 71.5540192 1.0000000
## 38-12 3.322860e+01 15.3209472 51.1362609 0.0000000
## 15-13 2.000000e+01 -127.1444030 167.1444030 1.0000000
## 16-13 5.000000e+01 -165.3973504 265.3973504 1.0000000
## 17-13 3.276515e+01 -91.9473609 157.4776640 0.9999999
## 18-13 6.666667e+01 -93.8810392 227.2143725 0.9993854
## 19-13 3.000000e+01 -117.1444030 177.1444030 1.0000000
## 20-13 4.314346e+01 -82.0008700 168.2877898 0.9999807
## 21-13 3.957851e+01 -84.8314033 163.9884244 0.9999962
## 22-13 4.629630e+01 -97.3019373 189.8945299 0.9999950
## 23-13 7.121212e+01 -63.9811457 206.4053882 0.9781456
## 27-13 -3.053202e-11 -215.3973504 215.3973504 1.0000000
## 28-13 1.631579e+01 -109.3461603 141.9777392 1.0000000
## 29-13 5.000000e+01 -165.3973504 265.3973504 1.0000000
## 30-13 3.333333e+01 -127.2143725 193.8810392 1.0000000
## 31-13 3.939394e+01 -88.6788244 167.4667032 0.9999981
## 32-13 1.000000e+02 -115.3973504 315.3973504 0.9962495
## 33-13 5.795455e+01 -81.0838463 196.9929373 0.9993441
## 34-13 1.000000e+02 -115.3973504 315.3973504 0.9962495
## 35-13 1.094236e-11 -147.1444030 147.1444030 1.0000000
## 38-13 4.105960e+01 -84.1209807 166.2401860 0.9999929
## 16-15 3.000000e+01 -162.6572470 222.6572470 1.0000000
## 17-15 1.276515e+01 -66.4434784 91.9737814 1.0000000
## 18-15 4.666667e+01 -81.7714980 175.1048314 0.9999460
## 19-15 1.000000e+01 -101.2307134 121.2307134 1.0000000
## 20-15 2.314346e+01 -56.7433325 103.0302523 0.9999995
## 21-15 1.957851e+01 -59.1528235 98.3098446 1.0000000
## 22-15 2.629630e+01 -80.1990040 132.7915966 1.0000000
## 23-15 5.121212e+01 -43.6457499 146.0699924 0.9703351
## 27-15 -2.000000e+01 -212.6572470 172.6572470 1.0000000
## 28-15 -3.684211e+00 -84.3794514 77.0110304 1.0000000
## 29-15 3.000000e+01 -162.6572470 222.6572470 1.0000000
## 30-15 1.333333e+01 -115.1048314 141.7714980 1.0000000
## 31-15 1.939394e+01 -65.0064700 103.7943488 1.0000000
## 32-15 8.000000e+01 -112.6572470 272.6572470 0.9993854
## 33-15 3.795455e+01 -62.3074647 138.2165556 0.9998828
## 34-15 8.000000e+01 -112.6572470 272.6572470 0.9993854
## 35-15 -2.000000e+01 -131.2307134 91.2307134 1.0000000
## 38-15 2.105960e+01 -58.8839696 101.0031749 0.9999999
## 17-16 -1.723485e+01 -193.3556885 158.8859915 1.0000000
## 18-16 1.666667e+01 -186.4119028 219.7452361 1.0000000
## 19-16 -2.000000e+01 -212.6572470 172.6572470 1.0000000
## 20-16 -6.856540e+00 -183.2834168 169.5703366 1.0000000
## 21-16 -1.042149e+01 -186.3281868 165.4852080 1.0000000
## 22-16 -3.703704e+00 -193.6663111 186.2589037 1.0000000
## 23-16 2.121212e+01 -162.4793564 204.9035988 1.0000000
## 27-16 -5.000000e+01 -298.7194364 198.7194364 1.0000000
## 28-16 -3.368421e+01 -210.4786254 143.1102043 1.0000000
## 29-16 2.806644e-12 -248.7194364 248.7194364 1.0000000
## 30-16 -1.666667e+01 -219.7452361 186.4119028 1.0000000
## 31-16 -1.060606e+01 -189.1220888 167.9099676 1.0000000
## 32-16 5.000000e+01 -198.7194364 298.7194364 1.0000000
## 33-16 7.954545e+00 -178.5850319 194.4941228 1.0000000
## 34-16 5.000000e+01 -198.7194364 298.7194364 1.0000000
## 35-16 -5.000000e+01 -242.6572470 142.6572470 1.0000000
## 38-16 -8.940397e+00 -185.3929914 167.5121967 1.0000000
## 18-17 3.390152e+01 -68.0695474 135.8725777 0.9999907
## 19-17 -2.765152e+00 -81.9737814 76.4434784 1.0000000
## 20-17 1.037831e+01 -6.4631594 27.2197762 0.8782856
## 21-17 6.813359e+00 -3.2045449 16.8312631 0.7306598
## 22-17 1.353114e+01 -58.8773107 85.9396002 1.0000000
## 23-17 3.844697e+01 -15.4023679 92.2963073 0.6326379
## 27-17 -3.276515e+01 -208.8859915 143.3556885 1.0000000
## 28-17 -1.644936e+01 -36.7830125 3.8842884 0.3510439
## 29-17 1.723485e+01 -158.8859915 193.3556885 1.0000000
## 30-17 5.681818e-01 -101.4028807 102.5392443 1.0000000
## 31-17 6.628788e+00 -25.3894031 38.6469788 1.0000000
## 32-17 6.723485e+01 -108.8859915 243.3556885 0.9998630
## 33-17 2.518939e+01 -37.6930847 88.0718726 0.9996721
## 34-17 6.723485e+01 -108.8859915 243.3556885 0.9998630
## 35-17 -3.276515e+01 -111.9737814 46.4434784 0.9994247
## 38-17 8.294451e+00 -8.8143238 25.4032261 0.9928843
## 19-18 -3.666667e+01 -165.1048314 91.7714980 0.9999996
## 20-18 -2.352321e+01 -126.0219391 78.9755256 1.0000000
## 21-18 -2.708816e+01 -128.6889115 74.5125993 0.9999999
## 22-18 -2.037037e+01 -144.7300886 103.9893478 1.0000000
## 23-18 4.545455e+00 -110.0063078 119.0972169 1.0000000
## 27-18 -6.666667e+01 -269.7452361 136.4119028 0.9999928
## 28-18 -5.035088e+01 -153.4809525 52.7791981 0.9921320
## 29-18 -1.666667e+01 -219.7452361 186.4119028 1.0000000
## 30-18 -3.333333e+01 -176.9315669 110.2649002 1.0000000
## 31-18 -2.727273e+01 -133.3270513 78.7815968 1.0000000
## 32-18 3.333333e+01 -169.7452361 236.4119028 1.0000000
## 33-18 -8.712121e+00 -127.7774865 110.3532441 1.0000000
## 34-18 3.333333e+01 -169.7452361 236.4119028 1.0000000
## 35-18 -6.666667e+01 -195.1048314 61.7714980 0.9818994
## 38-18 -2.560706e+01 -128.1500564 76.9359283 1.0000000
## 20-19 1.314346e+01 -66.7433325 93.0302523 1.0000000
## 21-19 9.578511e+00 -69.1528235 88.3098446 1.0000000
## 22-19 1.629630e+01 -90.1990040 122.7915966 1.0000000
## 23-19 4.121212e+01 -53.6457499 136.0699924 0.9986861
## 27-19 -3.000000e+01 -222.6572470 162.6572470 1.0000000
## 28-19 -1.368421e+01 -94.3794514 67.0110304 1.0000000
## 29-19 2.000000e+01 -172.6572470 212.6572470 1.0000000
## 30-19 3.333333e+00 -125.1048314 131.7714980 1.0000000
## 31-19 9.393939e+00 -75.0064700 93.7943488 1.0000000
## 32-19 7.000000e+01 -122.6572470 262.6572470 0.9999460
## 33-19 2.795455e+01 -72.3074647 128.2165556 0.9999998
## 34-19 7.000000e+01 -122.6572470 262.6572470 0.9999460
## 35-19 -3.000000e+01 -141.2307134 81.2307134 0.9999999
## 38-19 1.105960e+01 -68.8839696 91.0031749 1.0000000
## 21-20 -3.564949e+00 -17.9958518 10.8659531 1.0000000
## 22-20 3.152836e+00 -69.9968524 76.3025251 1.0000000
## 23-20 2.806866e+01 -26.7733272 82.9106498 0.9849488
## 27-20 -4.314346e+01 -219.5703366 133.2834168 1.0000000
## 28-20 -2.682767e+01 -49.6607580 -3.9945829 0.0040631
## 29-20 6.856540e+00 -169.5703366 183.2834168 1.0000000
## 30-20 -9.810127e+00 -112.3088589 92.6886058 1.0000000
## 31-20 -3.749521e+00 -37.4104340 29.9113929 1.0000000
## 32-20 5.685654e+01 -119.5703366 233.2834168 0.9999951
## 33-20 1.481109e+01 -48.9235100 78.5456811 1.0000000
## 34-20 5.685654e+01 -119.5703366 233.2834168 0.9999951
## 35-20 -4.314346e+01 -123.0302523 36.7433325 0.9702183
## 38-20 -2.083857e+00 -22.0989197 17.9312052 1.0000000
## 22-21 6.717786e+00 -65.1682374 78.6038089 1.0000000
## 23-21 3.163361e+01 -21.5111639 84.7783852 0.9126717
## 27-21 -3.957851e+01 -215.4852080 136.3281868 1.0000000
## 28-21 -2.326272e+01 -41.6494894 -4.8759528 0.0008921
## 29-21 1.042149e+01 -165.4852080 186.3281868 1.0000000
## 30-21 -6.245177e+00 -107.8459326 95.3555782 1.0000000
## 31-21 -1.845712e-01 -31.0030795 30.6339372 1.0000000
## 32-21 6.042149e+01 -115.4852080 236.3281868 0.9999820
## 33-21 1.837603e+01 -43.9041548 80.6562246 0.9999992
## 34-21 6.042149e+01 -115.4852080 236.3281868 0.9999820
## 35-21 -3.957851e+01 -118.3098446 39.1528235 0.9881783
## 38-21 1.481092e+00 -13.2608918 16.2230759 1.0000000
## 23-22 2.491582e+01 -64.3421892 114.1738390 0.9999998
## 27-22 -4.629630e+01 -236.2589037 143.6663111 1.0000000
## 28-22 -2.998051e+01 -104.0122518 44.0512382 0.9996024
## 29-22 3.703704e+00 -186.2589037 193.6663111 1.0000000
## 30-22 -1.296296e+01 -137.3226812 111.3967553 1.0000000
## 31-22 -6.902357e+00 -84.9562260 71.1515122 1.0000000
## 32-22 5.370370e+01 -136.2589037 243.6663111 0.9999997
## 33-22 1.165825e+01 -83.3230545 106.6395528 1.0000000
## 34-22 5.370370e+01 -136.2589037 243.6663111 0.9999997
## 35-22 -4.629630e+01 -152.7915966 60.1990040 0.9986729
## 38-22 -5.236694e+00 -78.4483874 67.9750001 1.0000000
## 27-23 -7.121212e+01 -254.9035988 112.4793564 0.9998182
## 28-23 -5.489633e+01 -110.9094196 1.1167561 0.0639274
## 29-23 -2.121212e+01 -204.9035988 162.4793564 1.0000000
## 30-23 -3.787879e+01 -152.4305502 76.6729745 0.9999917
## 31-23 -3.181818e+01 -93.0486744 29.4123107 0.9816311
## 32-23 2.878788e+01 -154.9035988 212.4793564 1.0000000
## 33-23 -1.325758e+01 -94.9779208 68.4627693 1.0000000
## 34-23 2.878788e+01 -154.9035988 212.4793564 1.0000000
## 35-23 -7.121212e+01 -166.0699924 23.6457499 0.5203264
## 38-23 -3.015252e+01 -85.0771837 24.7721466 0.9637754
## 28-27 1.631579e+01 -160.4786254 193.1102043 1.0000000
## 29-27 5.000000e+01 -198.7194364 298.7194364 1.0000000
## 30-27 3.333333e+01 -169.7452361 236.4119028 1.0000000
## 31-27 3.939394e+01 -139.1220888 217.9099676 1.0000000
## 32-27 1.000000e+02 -148.7194364 348.7194364 0.9996499
## 33-27 5.795455e+01 -128.5850319 244.4941228 0.9999977
## 34-27 1.000000e+02 -148.7194364 348.7194364 0.9996499
## 35-27 4.147438e-11 -192.6572470 192.6572470 1.0000000
## 38-27 4.105960e+01 -135.3929914 217.5121967 1.0000000
## 29-28 3.368421e+01 -143.1102043 210.4786254 1.0000000
## 30-28 1.701754e+01 -86.1125315 120.1476192 1.0000000
## 31-28 2.307815e+01 -12.4588646 58.6151644 0.8091312
## 32-28 8.368421e+01 -93.1102043 260.4786254 0.9949827
## 33-28 4.163876e+01 -23.1062900 106.3838019 0.8235922
## 34-28 8.368421e+01 -93.1102043 260.4786254 0.9949827
## 35-28 -1.631579e+01 -97.0110304 64.3794514 1.0000000
## 38-28 2.474381e+01 1.7128551 47.7747712 0.0185292
## 30-29 -1.666667e+01 -219.7452361 186.4119028 1.0000000
## 31-29 -1.060606e+01 -189.1220888 167.9099676 1.0000000
## 32-29 5.000000e+01 -198.7194364 298.7194364 1.0000000
## 33-29 7.954545e+00 -178.5850319 194.4941228 1.0000000
## 34-29 5.000000e+01 -198.7194364 298.7194364 1.0000000
## 35-29 -5.000000e+01 -242.6572470 142.6572470 1.0000000
## 38-29 -8.940397e+00 -185.3929914 167.5121967 1.0000000
## 31-30 6.060606e+00 -99.9937180 112.1149301 1.0000000
## 32-30 6.666667e+01 -136.4119028 269.7452361 0.9999928
## 33-30 2.462121e+01 -94.4441532 143.6865775 1.0000000
## 34-30 6.666667e+01 -136.4119028 269.7452361 0.9999928
## 35-30 -3.333333e+01 -161.7714980 95.1048314 1.0000000
## 38-30 7.726269e+00 -94.8167230 110.2692617 1.0000000
## 32-31 6.060606e+01 -117.9099676 239.1220888 0.9999858
## 33-31 1.856061e+01 -50.7476054 87.8688176 0.9999999
## 34-31 6.060606e+01 -117.9099676 239.1220888 0.9999858
## 35-31 -3.939394e+01 -123.7943488 45.0064700 0.9959326
## 38-31 1.665663e+00 -32.1297837 35.4611103 1.0000000
## 33-32 -4.204545e+01 -228.5850319 144.4941228 1.0000000
## 34-32 -1.607248e-11 -248.7194364 248.7194364 1.0000000
## 35-32 -1.000000e+02 -292.6572470 92.6572470 0.9818994
## 38-32 -5.894040e+01 -235.3929914 117.5121967 0.9999897
## 34-33 4.204545e+01 -144.4941228 228.5850319 1.0000000
## 35-33 -5.795455e+01 -158.2165556 42.3074647 0.9354421
## 38-33 -1.689494e+01 -80.7006934 46.9108078 0.9999999
## 35-34 -1.000000e+02 -292.6572470 92.6572470 0.9818994
## 38-34 -5.894040e+01 -235.3929914 117.5121967 0.9999897
## 38-35 4.105960e+01 -38.8839696 121.0031749 0.9842341# Visualize Variation in FatalityRate by Aircraft Category
ggplot(crash_data, aes(x = AirCraftCategory, y = FatalityRate)) +
geom_boxplot(fill = "lightblue", alpha = 0.6) +
coord_flip() + # Rotate for better readability
labs(title = "Variation of Fatality Rate by Aircraft Category",
x = "Aircraft Category", y = "Fatality Rate (%)") +
theme_minimal()## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
Interpretation:
The ANOVA results indicate whether there are significant differences in fatality rates across different aircraft categories.
The TukeyHSD post-hoc test identifies which specific categories differ significantly.
Q31. Does WeatherCondition significantly impact FatalityRate across multiple weather categories?
anova_model_weather <- aov(FatalityRate ~ WeatherCondition, data = crash_data)
anova_results_weather <- summary(anova_model_weather)
print(anova_results_weather)## Df Sum Sq Mean Sq F value Pr(>F)
## WeatherCondition 2 2146417 1073209 496.5 <2e-16 ***
## Residuals 20216 43701865 2162
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 24288 observations deleted due to missingnessTukeyHSD(anova_model_weather)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = FatalityRate ~ WeatherCondition, data = crash_data)
##
## $WeatherCondition
## diff lwr upr p adj
## Unknown-IMC -3.483057 -11.15653 4.190414 0.5365718
## VMC-IMC -35.079377 -37.80795 -32.350808 0.0000000
## VMC-Unknown -31.596320 -38.85849 -24.334154 0.0000000# Visualize Variation in FatalityRate by Weather Condition
ggplot(crash_data, aes(x = WeatherCondition, y = FatalityRate)) +
geom_boxplot(fill = "lightgreen", alpha = 0.6) +
coord_flip() + # Rotate for better readability
labs(title = "Variation of Fatality Rate by Weather Condition",
x = "Weather Condition", y = "Fatality Rate (%)") +
theme_minimal()## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
Interpretation:
The ANOVA results indicate whether there are significant differences in fatality rates across different weather conditions.
The TukeyHSD post-hoc test identifies which specific weather conditions differ significantly.
Q32. Does PurposeOfFlight significantly impact FatalityRate across multiple flight purposes?
table(crash_data$PurposeOfFlight) # Quick overview of categories##
## , ,, ,AOBV ,BUS ,EXEC
## 206 1 2 2 3
## ,FERY ,INST ,OWRK ,PERS ,POSI
## 2 3 2 14 1
## ,PUBF ,UNK ,UNK, AAPL AAPL,
## 1 4 1 1869 1
## AAPL,AAPL AAPL,FIRF AAPL,INST AAPL,PERS ADRP
## 18 1 1 1 12
## ADRP, AOBV AOBV, AOBV,AOBV AOBV,PERS
## 1 459 1 2 2
## ASHO ASHO,ASHO ASHO,INST BANT BUS
## 144 21 1 107 1159
## BUS, BUS,BUS BUS,INST EXEC EXEC,
## 2 7 2 200 2
## EXLD FERY FERY,PERS FIRF FIRF,FIRF
## 132 226 2 53 1
## FLTS FLTS, FLTS,FLTS GLDT GLDT,GLDT
## 462 1 1 53 1
## GLDT,PERS GLDT,PERS,GLDT INST INST, INST,EXEC
## 1 1 5176 1 1
## INST,GLDT INST,INST INST,OWRK INST,PERS INST,POSI
## 1 47 2 38 2
## INST,PUBF OWRK OWRK,INST OWRK,OWRK OWRK,PERS
## 1 680 2 6 3
## PERS PERS, PERS,AAPL PERS,AOBV PERS,ASHO
## 30667 12 3 2 1
## PERS,BUS PERS,GLDT PERS,INST PERS,OWRK PERS,PERS
## 3 2 29 2 165
## PERS,POSI PERS,PUBU PERS,SKYD PERS,UNK POSI
## 2 1 2 3 955
## POSI, POSI,AOBV POSI,BUS POSI,PERS POSI,POSI
## 1 1 1 4 5
## PUBF PUBL PUBL,PUBL PUBS PUBS,PERS
## 109 76 1 67 2
## PUBS,PUBS PUBU PUBU, PUBU,INST PUBU,PUBU
## 1 238 1 1 2
## SKYD SKYD,PERS SKYD,SKYD UNK UNK,
## 156 1 1 821 3
## UNK,INST UNK,PERS UNK,POSI UNK,UNK
## 2 4 1 10sort(table(crash_data$PurposeOfFlight), decreasing = TRUE) # Sorted by frequency##
## PERS INST AAPL BUS POSI
## 30667 5176 1869 1159 955
## UNK OWRK FLTS AOBV PUBU
## 821 680 462 459 238
## FERY , EXEC PERS,PERS SKYD
## 226 206 200 165 156
## ASHO EXLD PUBF BANT PUBL
## 144 132 109 107 76
## PUBS FIRF GLDT INST,INST INST,PERS
## 67 53 53 47 38
## PERS,INST ASHO,ASHO AAPL,AAPL ,PERS ADRP
## 29 21 18 14 12
## PERS, UNK,UNK BUS,BUS OWRK,OWRK POSI,POSI
## 12 10 7 6 5
## ,UNK POSI,PERS UNK,PERS ,EXEC ,INST
## 4 4 4 3 3
## OWRK,PERS PERS,AAPL PERS,BUS PERS,UNK UNK,
## 3 3 3 3 3
## ,AOBV ,BUS ,FERY ,OWRK AOBV,AOBV
## 2 2 2 2 2
## AOBV,PERS BUS, BUS,INST EXEC, FERY,PERS
## 2 2 2 2 2
## INST,OWRK INST,POSI OWRK,INST PERS,AOBV PERS,GLDT
## 2 2 2 2 2
## PERS,OWRK PERS,POSI PERS,SKYD PUBS,PERS PUBU,PUBU
## 2 2 2 2 2
## UNK,INST ,, ,POSI ,PUBF ,UNK,
## 2 1 1 1 1
## AAPL, AAPL,FIRF AAPL,INST AAPL,PERS ADRP,
## 1 1 1 1 1
## AOBV, ASHO,INST FIRF,FIRF FLTS, FLTS,FLTS
## 1 1 1 1 1
## GLDT,GLDT GLDT,PERS GLDT,PERS,GLDT INST, INST,EXEC
## 1 1 1 1 1
## INST,GLDT INST,PUBF PERS,ASHO PERS,PUBU POSI,
## 1 1 1 1 1
## POSI,AOBV POSI,BUS PUBL,PUBL PUBS,PUBS PUBU,
## 1 1 1 1 1
## PUBU,INST SKYD,PERS SKYD,SKYD UNK,POSI
## 1 1 1 1crash_data <- crash_data %>%
mutate(PurposeOfFlight_Clean = case_when(
str_detect(PurposeOfFlight, "PERS") ~ "Personal",
str_detect(PurposeOfFlight, "BUS") ~ "Business",
str_detect(PurposeOfFlight, "INST") ~ "Instructional",
str_detect(PurposeOfFlight, "POSI") ~ "Positioning",
str_detect(PurposeOfFlight, "FERY") ~ "Ferry",
str_detect(PurposeOfFlight, "AOBV|ASHO|ADRP|GLDT") ~ "Other",
str_detect(PurposeOfFlight, "MIL") ~ "Military",
str_detect(PurposeOfFlight, "PUBF|PUBL|PUBS|PUBU") ~ "Public Service",
str_detect(PurposeOfFlight, "SKYD") ~ "Skydiving",
TRUE ~ "Unknown"
))
crash_data$PurposeOfFlight_Clean <- as.factor(crash_data$PurposeOfFlight_Clean)
anova_model_purpose <- aov(FatalityRate ~ PurposeOfFlight_Clean, data = crash_data)
anova_results_purpose <- summary(anova_model_purpose)
print(anova_results_purpose)## Df Sum Sq Mean Sq F value Pr(>F)
## PurposeOfFlight_Clean 8 347833 43479 19.31 <2e-16 ***
## Residuals 20210 45500449 2251
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 24288 observations deleted due to missingnessTukeyHSD(anova_model_purpose)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = FatalityRate ~ PurposeOfFlight_Clean, data = crash_data)
##
## $PurposeOfFlight_Clean
## diff lwr upr p adj
## Ferry-Business -1.5793786 -17.1949411 14.03618395 0.9999973
## Instructional-Business -19.9410546 -27.2041998 -12.67790931 0.0000000
## Other-Business -6.7800366 -16.0892683 2.52919506 0.3672720
## Personal-Business -6.6757237 -13.0757403 -0.27570716 0.0332034
## Positioning-Business -0.3883593 -9.8919435 9.11522486 1.0000000
## Public Service-Business -16.0551990 -27.4451747 -4.66522332 0.0004203
## Skydiving-Business -19.6604051 -37.3719397 -1.94887044 0.0167802
## Unknown-Business -7.0995880 -14.1003153 -0.09886065 0.0438358
## Instructional-Ferry -18.3616760 -33.1154869 -3.60786503 0.0036258
## Other-Ferry -5.2006581 -21.0620405 10.66072439 0.9843777
## Personal-Ferry -5.0963452 -19.4449294 9.25223912 0.9741372
## Positioning-Ferry 1.1910193 -14.7852059 17.16724444 0.9999998
## Public Service-Ferry -14.4758204 -31.6412859 2.68964504 0.1797888
## Skydiving-Ferry -18.0810265 -39.9584148 3.79636178 0.2020961
## Unknown-Ferry -5.5202094 -20.1466185 9.10619965 0.9626115
## Other-Instructional 13.1610179 5.3834258 20.93861004 0.0000055
## Personal-Instructional 13.2653308 9.4196376 17.11102408 0.0000000
## Positioning-Instructional 19.5526953 11.5434970 27.56189348 0.0000000
## Public Service-Instructional 3.8858556 -6.2906218 14.06233287 0.9599260
## Skydiving-Instructional 0.2806495 -16.6759894 17.23728841 1.0000000
## Unknown-Instructional 12.8414666 8.0624157 17.62051748 0.0000000
## Personal-Other 0.1043129 -6.8740732 7.08269900 1.0000000
## Positioning-Other 6.3916773 -3.5106341 16.29398880 0.5416991
## Public Service-Other -9.2751624 -20.9998887 2.44956392 0.2549118
## Skydiving-Other -12.8803684 -30.8090081 5.04827121 0.3868814
## Unknown-Other -0.3195513 -7.8526678 7.21356511 1.0000000
## Positioning-Personal 6.2873644 -0.9482551 13.52298394 0.1492801
## Public Service-Personal -9.3794753 -18.9590172 0.20006668 0.0604520
## Skydiving-Personal -12.9846813 -29.5899376 3.62057497 0.2697985
## Unknown-Personal -0.4238642 -3.7474558 2.89972733 0.9999837
## Public Service-Positioning -15.6668397 -27.5464661 -3.78721331 0.0014258
## Skydiving-Positioning -19.2720458 -37.3023657 -1.24172586 0.0256961
## Unknown-Positioning -6.7112287 -14.4832400 1.06078266 0.1554044
## Skydiving-Public Service -3.6052061 -22.6972429 15.48683078 0.9996760
## Unknown-Public Service 8.9556110 -1.0352645 18.94648661 0.1210771
## Unknown-Skydiving 12.5608171 -4.2850877 29.40672187 0.3340770# Visualize Variation in FatalityRate by Purpose of Flight
ggplot(crash_data, aes(x = PurposeOfFlight_Clean, y = FatalityRate)) +
geom_boxplot(fill = "lightcoral", alpha = 0.6) +
coord_flip() + # Rotate for better readability
labs(title = "Variation of Fatality Rate by Purpose of Flight",
x = "Purpose of Flight", y = "Fatality Rate (%)") +
theme_minimal()## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
Interpretation:
The ANOVA results indicate whether there are significant differences in fatality rates across different flight purposes.
The TukeyHSD post-hoc test identifies which specific flight purposes differ significantly.
Q33. Is there a progressive increase in FatalityRate across IncidentSeverity levels?
crash_data$IncidentSeverity <- as.factor(crash_data$IncidentSeverity)
anova_model_severity <- aov(FatalityRate ~ IncidentSeverity, data = crash_data)
anova_results_severity <- summary(anova_model_severity)
print(anova_results_severity)## Df Sum Sq Mean Sq F value Pr(>F)
## IncidentSeverity 2 42220284 21110142 117630 <2e-16 ***
## Residuals 20216 3627998 179
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 24288 observations deleted due to missingnessTukeyHSD(anova_model_severity)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = FatalityRate ~ IncidentSeverity, data = crash_data)
##
## $IncidentSeverity
## diff lwr upr p adj
## Minor-Catastrophic -9.172854e+01 -92.2463829 -91.2106938 0
## Serious-Catastrophic -9.172854e+01 -92.2844971 -91.1725796 0
## Serious-Minor 1.738601e-11 -0.6032965 0.6032965 1# Visualize Variation in FatalityRate by Incident Severity
ggplot(crash_data, aes(x = IncidentSeverity, y = FatalityRate)) +
geom_boxplot(fill = "lightyellow", alpha = 0.6) +
coord_flip() + # Rotate for better readability
labs(title = "Variation of Fatality Rate by Incident Severity",
x = "Incident Severity", y = "Fatality Rate (%)") +
theme_minimal()## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
Interpretation:
The ANOVA results indicate whether there are significant differences in fatality rates across different incident severity levels.
The TukeyHSD post-hoc test identifies which specific severity levels differ significantly.
Q34. Does the severity level of AirCraftDamage correlate with higher FatalityRates?
crash_data$AirCraftDamage <- as.factor(crash_data$AirCraftDamage)
anova_model_damage <- aov(FatalityRate ~ AirCraftDamage, data = crash_data)
anova_results_damage <- summary(anova_model_damage)
print(anova_results_damage)## Df Sum Sq Mean Sq F value Pr(>F)
## AirCraftDamage 19 12612912 663837 403.5 <2e-16 ***
## Residuals 20199 33235370 1645
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 24288 observations deleted due to missingnessTukeyHSD(anova_model_damage)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = FatalityRate ~ AirCraftDamage, data = crash_data)
##
## $AirCraftDamage
## diff lwr
## Destroyed-, -1.914928e+01 -102.1760770
## Destroyed,Destroyed-, -1.021324e+01 -95.1711941
## Destroyed,Minor-, -1.848406e-10 -117.3862779
## Destroyed,None-, -5.000000e+01 -216.0092663
## Destroyed,Substantial-, -1.968391e+01 -106.8762463
## Minor-, -8.666940e+01 -170.1933535
## Minor,-, -1.104610e-10 -166.0092663
## Minor,Destroyed-, -3.333333e+01 -150.7196112
## Minor,Substantial-, -7.142857e+01 -170.6380836
## None-, -9.209115e+01 -175.2625179
## None,None-, -1.410427e-10 -166.0092663
## Substantial-, -7.251575e+01 -155.5297254
## Substantial,Destroyed-, -2.705882e+01 -117.0899889
## Substantial,Minor-, -5.769231e+01 -149.7776803
## Substantial,None-, -7.500000e+01 -184.8048085
## Substantial,Substantial-, -5.698148e+01 -141.3581850
## Substantial,Substantial,None-, -1.914628e-10 -166.0092663
## Unknown-, -3.691358e+01 -122.1927340
## Unknown,Unknown-, -1.916618e-10 -166.0092663
## Destroyed,Destroyed-Destroyed 8.936033e+00 -9.2783752
## Destroyed,Minor-Destroyed 1.914928e+01 -63.8775234
## Destroyed,None-Destroyed -3.085072e+01 -174.6317643
## Destroyed,Substantial-Destroyed -5.346312e-01 -27.3005642
## Minor-Destroyed -6.752012e+01 -77.0155328
## Minor,-Destroyed 1.914928e+01 -124.6317643
## Minor,Destroyed-Destroyed -1.418406e+01 -97.2108567
## Minor,Substantial-Destroyed -5.227929e+01 -106.6524371
## None-Destroyed -7.294188e+01 -78.5443111
## None,None-Destroyed 1.914928e+01 -124.6317643
## Substantial-Destroyed -5.336647e+01 -55.6536917
## Substantial,Destroyed-Destroyed -7.909547e+00 -42.8312025
## Substantial,Minor-Destroyed -3.854303e+01 -78.4632908
## Substantial,None-Destroyed -5.585072e+01 -127.7604393
## Substantial,Substantial-Destroyed -3.783220e+01 -53.1076557
## Substantial,Substantial,None-Destroyed 1.914928e+01 -124.6317643
## Unknown-Destroyed -1.776430e+01 -37.4225176
## Unknown,Unknown-Destroyed 1.914928e+01 -124.6317643
## Destroyed,Minor-Destroyed,Destroyed 1.021324e+01 -74.7447070
## Destroyed,None-Destroyed,Destroyed -3.978676e+01 -184.6915238
## Destroyed,Substantial-Destroyed,Destroyed -9.470664e+00 -41.7323907
## Minor-Destroyed,Destroyed -7.645615e+01 -96.8170630
## Minor,-Destroyed,Destroyed 1.021324e+01 -134.6915238
## Minor,Destroyed-Destroyed,Destroyed -2.312009e+01 -108.0780404
## Minor,Substantial-Destroyed,Destroyed -6.121533e+01 -118.4939665
## None-Destroyed,Destroyed -8.187791e+01 -100.7403346
## None,None-Destroyed,Destroyed 1.021324e+01 -134.6915238
## Substantial-Destroyed,Destroyed -6.230251e+01 -80.4583647
## Substantial,Destroyed-Destroyed,Destroyed -1.684558e+01 -56.1384019
## Substantial,Minor-Destroyed,Destroyed -4.747906e+01 -91.2743945
## Substantial,None-Destroyed,Destroyed -6.478676e+01 -138.9177964
## Substantial,Substantial-Destroyed,Destroyed -4.676824e+01 -70.3848256
## Substantial,Substantial,None-Destroyed,Destroyed 1.021324e+01 -134.6915238
## Unknown-Destroyed,Destroyed -2.670034e+01 -53.3620920
## Unknown,Unknown-Destroyed,Destroyed 1.021324e+01 -134.6915238
## Destroyed,None-Destroyed,Minor -5.000000e+01 -216.0092663
## Destroyed,Substantial-Destroyed,Minor -1.968391e+01 -106.8762463
## Minor-Destroyed,Minor -8.666940e+01 -170.1933535
## Minor,-Destroyed,Minor 7.437961e-11 -166.0092663
## Minor,Destroyed-Destroyed,Minor -3.333333e+01 -150.7196112
## Minor,Substantial-Destroyed,Minor -7.142857e+01 -170.6380836
## None-Destroyed,Minor -9.209115e+01 -175.2625179
## None,None-Destroyed,Minor 4.379785e-11 -166.0092663
## Substantial-Destroyed,Minor -7.251575e+01 -155.5297254
## Substantial,Destroyed-Destroyed,Minor -2.705882e+01 -117.0899889
## Substantial,Minor-Destroyed,Minor -5.769231e+01 -149.7776803
## Substantial,None-Destroyed,Minor -7.500000e+01 -184.8048085
## Substantial,Substantial-Destroyed,Minor -5.698148e+01 -141.3581850
## Substantial,Substantial,None-Destroyed,Minor -6.622258e-12 -166.0092663
## Unknown-Destroyed,Minor -3.691358e+01 -122.1927340
## Unknown,Unknown-Destroyed,Minor -6.821210e-12 -166.0092663
## Destroyed,Substantial-Destroyed,None 3.031609e+01 -115.9099047
## Minor-Destroyed,None -3.666940e+01 -180.7380937
## Minor,-Destroyed,None 5.000000e+01 -153.3189974
## Minor,Destroyed-Destroyed,None 1.666667e+01 -149.3425996
## Minor,Substantial-Destroyed,None -2.142857e+01 -175.1232869
## None-Destroyed,None -4.209115e+01 -185.9557218
## None,None-Destroyed,None 5.000000e+01 -153.3189974
## Substantial-Destroyed,None -2.251575e+01 -166.2893859
## Substantial,Destroyed-Destroyed,None 2.294118e+01 -124.9951239
## Substantial,Minor-Destroyed,None -7.692308e+00 -156.8876631
## Substantial,None-Destroyed,None -2.500000e+01 -185.7377809
## Substantial,Substantial-Destroyed,None -6.981481e+00 -151.5462294
## Substantial,Substantial,None-Destroyed,None 5.000000e+01 -153.3189974
## Unknown-Destroyed,None 1.308642e+01 -132.0069030
## Unknown,Unknown-Destroyed,None 5.000000e+01 -153.3189974
## Minor-Destroyed,Substantial -6.698549e+01 -95.2559171
## Minor,-Destroyed,Substantial 1.968391e+01 -126.5420886
## Minor,Destroyed-Destroyed,Substantial -1.364943e+01 -100.8417636
## Minor,Substantial-Destroyed,Substantial -5.174466e+01 -112.2879779
## None-Destroyed,Substantial -7.240724e+01 -99.6183010
## None,None-Destroyed,Substantial 1.968391e+01 -126.5420886
## Substantial-Destroyed,Substantial -5.283184e+01 -79.5579657
## Substantial,Destroyed-Destroyed,Substantial -7.374915e+00 -51.2904748
## Substantial,Minor-Destroyed,Substantial -3.800840e+01 -85.9946629
## Substantial,None-Destroyed,Substantial -5.531609e+01 -131.9976515
## Substantial,Substantial-Destroyed,Substantial -3.729757e+01 -67.9960051
## Substantial,Substantial,None-Destroyed,Substantial 1.968391e+01 -126.5420886
## Unknown-Destroyed,Substantial -1.722967e+01 -50.3280058
## Unknown,Unknown-Destroyed,Substantial 1.968391e+01 -126.5420886
## Minor,-Minor 8.666940e+01 -57.3993028
## Minor,Destroyed-Minor 5.333606e+01 -30.1878960
## Minor,Substantial-Minor 1.524082e+01 -39.8884845
## None-Minor -5.421757e+00 -16.1076969
## None,None-Minor 8.666940e+01 -57.3993028
## Substantial-Minor 1.415365e+01 4.7710340
## Substantial,Destroyed-Minor 5.961057e+01 23.5228445
## Substantial,Minor-Minor 2.897709e+01 -11.9671326
## Substantial,None-Minor 1.166940e+01 -60.8137702
## Substantial,Substantial-Minor 2.968791e+01 11.9075476
## Substantial,Substantial,None-Minor 8.666940e+01 -57.3993028
## Unknown-Minor 4.975582e+01 28.0936960
## Unknown,Unknown-Minor 8.666940e+01 -57.3993028
## Minor,Destroyed-Minor, -3.333333e+01 -199.3425996
## Minor,Substantial-Minor, -7.142857e+01 -225.1232869
## None-Minor, -9.209115e+01 -235.9557218
## None,None-Minor, -3.058176e-11 -203.3189974
## Substantial-Minor, -7.251575e+01 -216.2893859
## Substantial,Destroyed-Minor, -2.705882e+01 -174.9951239
## Substantial,Minor-Minor, -5.769231e+01 -206.8876631
## Substantial,None-Minor, -7.500000e+01 -235.7377809
## Substantial,Substantial-Minor, -5.698148e+01 -201.5462294
## Substantial,Substantial,None-Minor, -8.100187e-11 -203.3189974
## Unknown-Minor, -3.691358e+01 -182.0069030
## Unknown,Unknown-Minor, -8.120082e-11 -203.3189974
## Minor,Substantial-Minor,Destroyed -3.809524e+01 -137.3047503
## None-Minor,Destroyed -5.875782e+01 -141.9291845
## None,None-Minor,Destroyed 3.333333e+01 -132.6759329
## Substantial-Minor,Destroyed -3.918242e+01 -122.1963920
## Substantial,Destroyed-Minor,Destroyed 6.274510e+00 -83.7566555
## Substantial,Minor-Minor,Destroyed -2.435897e+01 -116.4443469
## Substantial,None-Minor,Destroyed -4.166667e+01 -151.4714751
## Substantial,Substantial-Minor,Destroyed -2.364815e+01 -108.0248516
## Substantial,Substantial,None-Minor,Destroyed 3.333333e+01 -132.6759329
## Unknown-Minor,Destroyed -3.580247e+00 -88.8594007
## Unknown,Unknown-Minor,Destroyed 3.333333e+01 -132.6759329
## None-Minor,Substantial -2.066258e+01 -75.2562169
## None,None-Minor,Substantial 7.142857e+01 -82.2661440
## Substantial-Minor,Substantial -1.087179e+00 -55.4407356
## Substantial,Destroyed-Minor,Substantial 4.436975e+01 -20.1949468
## Substantial,Minor-Minor,Substantial 1.373626e+01 -53.6633276
## Substantial,None-Minor,Substantial -3.571429e+00 -93.6829430
## Substantial,Substantial-Minor,Substantial 1.444709e+01 -41.9658267
## Substantial,Substantial,None-Minor,Substantial 7.142857e+01 -82.2661440
## Unknown-Minor,Substantial 3.451499e+01 -23.2389970
## Unknown,Unknown-Minor,Substantial 7.142857e+01 -82.2661440
## None,None-None 9.209115e+01 -51.7734162
## Substantial-None 1.957540e+01 14.1663563
## Substantial,Destroyed-None 6.503233e+01 29.7683469
## Substantial,Minor-None 3.439885e+01 -5.8212191
## Substantial,None-None 1.709115e+01 -54.9854293
## Substantial,Substantial-None 3.510967e+01 19.0670463
## Substantial,Substantial,None-None 9.209115e+01 -51.7734162
## Unknown-None 5.517757e+01 34.9174690
## Unknown,Unknown-None 9.209115e+01 -51.7734162
## Substantial-None,None -7.251575e+01 -216.2893859
## Substantial,Destroyed-None,None -2.705882e+01 -174.9951239
## Substantial,Minor-None,None -5.769231e+01 -206.8876631
## Substantial,None-None,None -7.500000e+01 -235.7377809
## Substantial,Substantial-None,None -5.698148e+01 -201.5462294
## Substantial,Substantial,None-None,None -5.042011e-11 -203.3189974
## Unknown-None,None -3.691358e+01 -182.0069030
## Unknown,Unknown-None,None -5.061906e-11 -203.3189974
## Substantial,Destroyed-Substantial 4.545693e+01 10.5657740
## Substantial,Minor-Substantial 1.482344e+01 -25.0701365
## Substantial,None-Substantial -2.484250e+00 -74.3791575
## Substantial,Substantial-Substantial 1.553427e+01 0.3286809
## Substantial,Substantial,None-Substantial 7.251575e+01 -71.2578853
## Unknown-Substantial 3.560217e+01 15.9981938
## Unknown,Unknown-Substantial 7.251575e+01 -71.2578853
## Substantial,Minor-Substantial,Destroyed -3.063348e+01 -83.6031767
## Substantial,None-Substantial,Destroyed -4.794118e+01 -127.8359100
## Substantial,Substantial-Substantial,Destroyed -2.992266e+01 -67.9423973
## Substantial,Substantial,None-Substantial,Destroyed 2.705882e+01 -120.8774769
## Unknown-Substantial,Destroyed -9.854757e+00 -49.8373349
## Unknown,Unknown-Substantial,Destroyed 2.705882e+01 -120.8774769
## Substantial,None-Substantial,Minor -1.730769e+01 -99.5103295
## Substantial,Substantial-Substantial,Minor 7.108262e-01 -41.9460102
## Substantial,Substantial,None-Substantial,Minor 5.769231e+01 -91.5030477
## Unknown-Substantial,Minor 2.077873e+01 -23.6364915
## Unknown,Unknown-Substantial,Minor 5.769231e+01 -91.5030477
## Substantial,Substantial-Substantial,None 1.801852e+01 -55.4456619
## Substantial,Substantial,None-Substantial,None 7.500000e+01 -85.7377809
## Unknown-Substantial,None 3.808642e+01 -36.4125184
## Unknown,Unknown-Substantial,None 7.500000e+01 -85.7377809
## Substantial,Substantial,None-Substantial,Substantial 5.698148e+01 -87.5832664
## Unknown-Substantial,Substantial 2.006790e+01 -4.6792990
## Unknown,Unknown-Substantial,Substantial 5.698148e+01 -87.5832664
## Unknown-Substantial,Substantial,None -3.691358e+01 -182.0069030
## Unknown,Unknown-Substantial,Substantial,None -1.989520e-13 -203.3189974
## Unknown,Unknown-Unknown 3.691358e+01 -108.1797425
## upr p adj
## Destroyed-, 63.87752339 0.9999994
## Destroyed,Destroyed-, 74.74470702 1.0000000
## Destroyed,Minor-, 117.38627791 1.0000000
## Destroyed,None-, 116.00926625 0.9999567
## Destroyed,Substantial-, 67.50843025 0.9999996
## Minor-, -3.14543732 0.0319149
## Minor,-, 166.00926625 1.0000000
## Minor,Destroyed-, 84.05294457 0.9999826
## Minor,Substantial-, 27.78094079 0.5429617
## None-, -8.91978775 0.0130061
## None,None-, 166.00926625 1.0000000
## Substantial-, 10.49822475 0.1833122
## Substantial,Destroyed-, 62.97234180 0.9999581
## Substantial,Minor-, 34.39306487 0.7880275
## Substantial,None-, 34.80480846 0.6455960
## Substantial,Substantial-, 27.39522200 0.6664419
## Substantial,Substantial,None-, 166.00926625 1.0000000
## Unknown-, 48.36557354 0.9935625
## Unknown,Unknown-, 166.00926625 1.0000000
## Destroyed,Destroyed-Destroyed 27.15044174 0.9738704
## Destroyed,Minor-Destroyed 102.17607702 0.9999994
## Destroyed,None-Destroyed 112.93031796 0.9999998
## Destroyed,Substantial-Destroyed 26.23130176 1.0000000
## Minor-Destroyed -58.02470446 0.0000000
## Minor,-Destroyed 162.93031796 1.0000000
## Minor,Destroyed-Destroyed 68.84274369 1.0000000
## Minor,Substantial-Destroyed 2.09384790 0.0769184
## None-Destroyed -67.33944094 0.0000000
## None,None-Destroyed 162.93031796 1.0000000
## Substantial-Destroyed -51.07925532 0.0000000
## Substantial,Destroyed-Destroyed 27.01210906 0.9999996
## Substantial,Minor-Destroyed 1.37722905 0.0736466
## Substantial,None-Destroyed 16.05899294 0.3891094
## Substantial,Substantial-Destroyed -22.55675368 0.0000000
## Substantial,Substantial,None-Destroyed 162.93031796 1.0000000
## Unknown-Destroyed 1.89391071 0.1387390
## Unknown,Unknown-Destroyed 162.93031796 1.0000000
## Destroyed,Minor-Destroyed,Destroyed 95.17119411 1.0000000
## Destroyed,None-Destroyed,Destroyed 105.11801091 0.9999898
## Destroyed,Substantial-Destroyed,Destroyed 22.79106175 0.9999708
## Minor-Destroyed,Destroyed -56.09524077 0.0000000
## Minor,-Destroyed,Destroyed 155.11801091 1.0000000
## Minor,Destroyed-Destroyed,Destroyed 61.83786078 0.9999911
## Minor,Substantial-Destroyed,Destroyed -3.93668930 0.0216047
## None-Destroyed,Destroyed -63.01548392 0.0000000
## None,None-Destroyed,Destroyed 155.11801091 1.0000000
## Substantial-Destroyed,Destroyed -44.14664879 0.0000000
## Substantial,Destroyed-Destroyed,Destroyed 22.44724192 0.9942645
## Substantial,Minor-Destroyed,Destroyed -3.68373376 0.0176958
## Substantial,None-Destroyed,Destroyed 9.34428350 0.1826393
## Substantial,Substantial-Destroyed,Destroyed -23.15165024 0.0000000
## Substantial,Substantial,None-Destroyed,Destroyed 155.11801091 1.0000000
## Unknown-Destroyed,Destroyed -0.03858141 0.0491693
## Unknown,Unknown-Destroyed,Destroyed 155.11801091 1.0000000
## Destroyed,None-Destroyed,Minor 116.00926625 0.9999567
## Destroyed,Substantial-Destroyed,Minor 67.50843025 0.9999996
## Minor-Destroyed,Minor -3.14543732 0.0319149
## Minor,-Destroyed,Minor 166.00926625 1.0000000
## Minor,Destroyed-Destroyed,Minor 84.05294457 0.9999826
## Minor,Substantial-Destroyed,Minor 27.78094079 0.5429617
## None-Destroyed,Minor -8.91978775 0.0130061
## None,None-Destroyed,Minor 166.00926625 1.0000000
## Substantial-Destroyed,Minor 10.49822475 0.1833122
## Substantial,Destroyed-Destroyed,Minor 62.97234180 0.9999581
## Substantial,Minor-Destroyed,Minor 34.39306487 0.7880275
## Substantial,None-Destroyed,Minor 34.80480846 0.6455960
## Substantial,Substantial-Destroyed,Minor 27.39522200 0.6664419
## Substantial,Substantial,None-Destroyed,Minor 166.00926625 1.0000000
## Unknown-Destroyed,Minor 48.36557354 0.9935625
## Unknown,Unknown-Destroyed,Minor 166.00926625 1.0000000
## Destroyed,Substantial-Destroyed,None 176.54208861 0.9999999
## Minor-Destroyed,None 107.39930283 0.9999970
## Minor,-Destroyed,None 253.31899744 0.9999983
## Minor,Destroyed-Destroyed,None 182.67593292 1.0000000
## Minor,Substantial-Destroyed,None 132.26614402 1.0000000
## None-Destroyed,None 101.77341616 0.9999723
## None,None-Destroyed,None 253.31899744 0.9999983
## Substantial-Destroyed,None 121.25788530 1.0000000
## Substantial,Destroyed-Destroyed,None 170.87747685 1.0000000
## Substantial,Minor-Destroyed,None 141.50304773 1.0000000
## Substantial,None-Destroyed,None 135.73778088 1.0000000
## Substantial,Substantial-Destroyed,None 137.58326641 1.0000000
## Substantial,Substantial,None-Destroyed,None 253.31899744 0.9999983
## Unknown-Destroyed,None 158.17974252 1.0000000
## Unknown,Unknown-Destroyed,None 253.31899744 0.9999983
## Minor-Destroyed,Substantial -38.71505770 0.0000000
## Minor,-Destroyed,Substantial 165.90990470 1.0000000
## Minor,Destroyed-Destroyed,Substantial 73.54291300 1.0000000
## Minor,Substantial-Destroyed,Substantial 8.79865112 0.2159906
## None-Destroyed,Substantial -45.19618851 0.0000000
## None,None-Destroyed,Substantial 165.90990470 1.0000000
## Substantial-Destroyed,Substantial -26.10571888 0.0000000
## Substantial,Destroyed-Destroyed,Substantial 36.54064379 1.0000000
## Substantial,Minor-Destroyed,Substantial 9.97786356 0.3505330
## Substantial,None-Destroyed,Substantial 21.36546761 0.5390517
## Substantial,Substantial-Destroyed,Substantial -6.59914176 0.0027581
## Substantial,Substantial,None-Destroyed,Substantial 165.90990470 1.0000000
## Unknown-Destroyed,Substantial 15.86866141 0.9528713
## Unknown,Unknown-Destroyed,Substantial 165.90990470 1.0000000
## Minor,-Minor 230.73809369 0.8403321
## Minor,Destroyed-Minor 136.86002020 0.7600646
## Minor,Substantial-Minor 70.37013247 0.9999886
## None-Minor 5.26418216 0.9632678
## None,None-Minor 230.73809369 0.8403321
## Substantial-Minor 23.53625629 0.0000168
## Substantial,Destroyed-Minor 95.69829927 0.0000009
## Substantial,Minor-Minor 69.92130806 0.5772696
## Substantial,None-Minor 84.15256104 1.0000000
## Substantial,Substantial-Minor 47.46828032 0.0000006
## Substantial,Substantial,None-Minor 230.73809369 0.8403321
## Unknown-Minor 71.41793439 0.0000000
## Unknown,Unknown-Minor 230.73809369 0.8403321
## Minor,Destroyed-Minor, 132.67593292 0.9999999
## Minor,Substantial-Minor, 82.26614402 0.9853855
## None-Minor, 51.77341616 0.7563539
## None,None-Minor, 203.31899744 1.0000000
## Substantial-Minor, 71.25788530 0.9653692
## Substantial,Destroyed-Minor, 120.87747685 1.0000000
## Substantial,Minor-Minor, 91.50304773 0.9984463
## Substantial,None-Minor, 85.73778088 0.9847270
## Substantial,Substantial-Minor, 87.58326641 0.9980034
## Substantial,Substantial,None-Minor, 203.31899744 1.0000000
## Unknown-Minor, 108.17974252 0.9999970
## Unknown,Unknown-Minor, 203.31899744 1.0000000
## Minor,Substantial-Minor,Destroyed 61.11427412 0.9985840
## None-Minor,Destroyed 24.41354558 0.5807751
## None,None-Minor,Destroyed 199.34259958 0.9999999
## Substantial-Minor,Destroyed 43.83155808 0.9826741
## Substantial,Destroyed-Minor,Destroyed 96.30567513 1.0000000
## Substantial,Minor-Minor,Destroyed 67.72639821 0.9999944
## Substantial,None-Minor,Destroyed 68.13814179 0.9987912
## Substantial,Substantial-Minor,Destroyed 60.72855534 0.9999859
## Substantial,Substantial,None-Minor,Destroyed 199.34259958 0.9999999
## Unknown-Minor,Destroyed 81.69890688 1.0000000
## Unknown,Unknown-Minor,Destroyed 199.34259958 0.9999999
## None-Minor,Substantial 33.93105415 0.9988324
## None,None-Minor,Substantial 225.12328687 0.9853855
## Substantial-Minor,Substantial 53.26637785 1.0000000
## Substantial,Destroyed-Minor,Substantial 108.93444262 0.6341606
## Substantial,Minor-Minor,Substantial 81.13585509 0.9999999
## Substantial,None-Minor,Substantial 86.54008587 1.0000000
## Substantial,Substantial-Minor,Substantial 70.86000660 0.9999967
## Substantial,Substantial,None-Minor,Substantial 225.12328687 0.9853855
## Unknown-Minor,Substantial 92.26897932 0.8479123
## Unknown,Unknown-Minor,Substantial 225.12328687 0.9853855
## None,None-None 235.95572179 0.7563539
## Substantial-None 24.98444873 0.0000000
## Substantial,Destroyed-None 100.29631170 0.0000000
## Substantial,Minor-None 74.61890935 0.2149047
## Substantial,None-None 89.16773491 0.9999991
## Substantial,Substantial-None 51.15229634 0.0000000
## Substantial,Substantial,None-None 235.95572179 0.7563539
## Unknown-None 75.43767616 0.0000000
## Unknown,Unknown-None 235.95572179 0.7563539
## Substantial-None,None 71.25788530 0.9653692
## Substantial,Destroyed-None,None 120.87747685 1.0000000
## Substantial,Minor-None,None 91.50304773 0.9984463
## Substantial,None-None,None 85.73778088 0.9847270
## Substantial,Substantial-None,None 87.58326641 0.9980034
## Substantial,Substantial,None-None,None 203.31899744 1.0000000
## Unknown-None,None 108.17974252 0.9999970
## Unknown,Unknown-None,None 203.31899744 1.0000000
## Substantial,Destroyed-Substantial 80.34807959 0.0006811
## Substantial,Minor-Substantial 54.71702173 0.9990896
## Substantial,None-Substantial 69.41065816 1.0000000
## Substantial,Substantial-Substantial 30.73985680 0.0387683
## Substantial,Substantial,None-Substantial 216.28938592 0.9653692
## Unknown-Substantial 55.20614634 0.0000000
## Unknown,Unknown-Substantial 216.28938592 0.9653692
## Substantial,Minor-Substantial,Destroyed 22.33620839 0.8817763
## Substantial,None-Substantial,Destroyed 31.95355708 0.8432826
## Substantial,Substantial-Substantial,Destroyed 8.09708140 0.3629578
## Substantial,Substantial,None-Substantial,Destroyed 174.99512391 1.0000000
## Unknown-Substantial,Destroyed 30.12782150 0.9999982
## Unknown,Unknown-Substantial,Destroyed 174.99512391 1.0000000
## Substantial,None-Substantial,Minor 64.89494488 0.9999999
## Substantial,Substantial-Substantial,Minor 43.36766260 1.0000000
## Substantial,Substantial,None-Substantial,Minor 206.88766312 0.9984463
## Unknown-Substantial,Minor 65.19394636 0.9842772
## Unknown,Unknown-Substantial,Minor 206.88766312 0.9984463
## Substantial,Substantial-Substantial,None 91.48269898 0.9999984
## Substantial,Substantial,None-Substantial,None 235.73778088 0.9847270
## Unknown-Substantial,None 112.58535794 0.9604217
## Unknown,Unknown-Substantial,None 235.73778088 0.9847270
## Substantial,Substantial,None-Substantial,Substantial 201.54622937 0.9980034
## Unknown-Substantial,Substantial 44.81510152 0.3060106
## Unknown,Unknown-Substantial,Substantial 201.54622937 0.9980034
## Unknown-Substantial,Substantial,None 108.17974252 0.9999970
## Unknown,Unknown-Substantial,Substantial,None 203.31899744 1.0000000
## Unknown,Unknown-Unknown 182.00690302 0.9999970# Visualize Variation in FatalityRate by Aircraft Damage
ggplot(crash_data, aes(x = AirCraftDamage, y = FatalityRate)) +
geom_boxplot(fill = "lightpink", alpha = 0.6) +
coord_flip() + # Rotate for better readability
labs(title = "Variation of Fatality Rate by Aircraft Damage",
x = "Aircraft Damage", y = "Fatality Rate (%)") +
theme_minimal()## Warning: Removed 24288 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
Interpretation:
The ANOVA results indicate whether there are significant differences in fatality rates across different aircraft damage levels.
The TukeyHSD post-hoc test identifies which specific damage levels differ significantly.
Conclusion
🔹 Data Preprocessing
- Cleaned the aviation dataset by handling missing values and standardizing important fields.
- Created new features like
FatalityRateand categorized incidents byIncidentSeverity.
🔹 Key Findings from EDA
- Aviation incidents are concentrated in a few specific states.
- Small aircraft categories (like light airplanes) show higher average fatality rates.
- Certain months have higher incident rates, suggesting seasonal patterns.
🔹 Regression Insights
- Fatality rate is influenced by multiple factors like Number of Engines, Weather Condition, and Latitude.
- Incidents involving amateur-built aircraft have a statistically higher fatality risk.
- Fatalities have decreased over time, but occasional spikes indicate areas for improvement.
🔹 ANOVA Test Results
- Significant differences in
FatalityRateexist across AirCraftCategory, WeatherCondition, PurposeOfFlight, and IncidentSeverity. - Damage severity (
AirCraftDamage) is strongly linked to higher fatality rates.
🔹 Geographical Impact
- State-wise analysis shows accident hotspots.
- Incidents in the Southern Hemisphere showed higher fatal injury counts compared to the Northern Hemisphere.
🔹 Temporal Trend
- There is an overall decline in fatal injuries over the years, reflecting gradual improvements in aviation safety.
🔹 Engine Count
- The number of engines slightly affects the severity — but it’s less impactful compared to weather and flight purpose.
🔹 Weather Influence
- Poor weather conditions (IMC) drastically increase fatality rates compared to visual conditions (VMC).
Summary
Overall, multiple operational and environmental factors — including aircraft type, flight purpose, weather conditions, and damage levels — significantly impact the severity of aviation accidents.
This analysis supports data-driven recommendations for targeted safety improvements in the aviation sector.
Key Takeaways and Insights
Aircraft Category Matters:
Certain aircraft categories, especially smaller planes, have higher fatality risks compared to others.Flight Purpose Impact:
Personal and non-commercial flights are riskier than commercial or military flights.Weather Conditions are Critical:
Poor visibility conditions (IMC) significantly increase the fatality rate compared to visual conditions (VMC).Aircraft Damage Severity Reflects Fatality Risk:
Greater physical damage to aircraft correlates strongly with higher fatality rates.Geographical Variations:
Differences across states and between the Northern and Southern hemispheres impact crash outcomes.Engine Count Influence:
The number of engines slightly affects the fatality rate, but less than operational or environmental factors.Multi-Factor Fatality Prediction:
FatalityRate is better explained when combining variables like WeatherCondition, NumberOfEngines, and PurposeOfFlight.Temporal Trends:
While safety has improved over the years, occasional spikes in fatalities highlight areas still needing attention.
Summary:
The project highlights that aviation safety is a multi-dimensional challenge, influenced by aircraft type, environmental conditions, flight purpose, and operational settings.
Proactive, data-driven interventions targeting these factors can substantially improve aviation outcomes.
REFERENCES
- National Transportation Safety Board (NTSB) Aviation Accident Database