School Shooting Data Analysis (1)
2026-10-20
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.2.1 ✔ readr 2.2.0
## ✔ forcats 1.0.1 ✔ stringr 1.6.0
## ✔ ggplot2 4.0.3 ✔ tibble 3.3.1
## ✔ lubridate 1.9.5 ✔ tidyr 1.3.2
## ✔ purrr 1.2.2
## ── 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 errorsk12 <- read_csv("Violence Project - K-12 SCHOOLS DATABASE 2026 - Event.csv")## Rows: 398 Columns: 58
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (46): Full Date, Day of Week, Time of Day, During Class?, School Name, ...
## dbl (10): ID #, Month, Day, Year, Latitude, Longitude, Number of
## Victims Ki...
## num (1): Enrollment
## time (1): Time Homicide
## Occurred
##
## 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.higher_ed <- read_csv("Violence Project - HIGHER ED DATABASE 2026 - Event.csv")## Rows: 280 Columns: 60
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (48): Full Date, Day of Week, Time of Day, During Class?, School Name, ...
## dbl (10): ID #, Month, Day, Year, Latitude, Longitude, Number of
## Victims Ki...
## num (1): Enrollment
## time (1): Time Homicide
## Occurred
##
## 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(k12)
View(higher_ed)k12 <- k12 %>%
mutate(School_Type_Group = "K-12")
higher_ed <- higher_ed %>%
mutate(School_Type_Group = "Higher Ed")View(k12)
View(higher_ed)# Columns found only in K-12
setdiff(names(k12), names(higher_ed))## [1] "School Level"# Columns found only in higher education
setdiff(names(higher_ed), names(k12))## [1] "School Type" "Endowment (in millions)"
## [3] "HBCU"all_shootings <- bind_rows(k12, higher_ed)library(janitor)##
## Attaching package: 'janitor'## The following objects are masked from 'package:stats':
##
## chisq.test, fisher.testall_shootings <- all_shootings %>%
clean_names()names(all_shootings)## [1] "id_number"
## [2] "full_date"
## [3] "month"
## [4] "day"
## [5] "year"
## [6] "day_of_week"
## [7] "time_of_day"
## [8] "time_homicide_occurred"
## [9] "during_class"
## [10] "school_name"
## [11] "address"
## [12] "city"
## [13] "state"
## [14] "zip_code"
## [15] "density"
## [16] "latitude"
## [17] "longitude"
## [18] "industry_type_of_business"
## [19] "school_level"
## [20] "number_of_employees_at_site"
## [21] "enrollment"
## [22] "accessibility_to_public"
## [23] "gun_free_zone"
## [24] "security_systems"
## [25] "lockdown"
## [26] "how_perpetrator_accessed"
## [27] "armed_person_on_scene"
## [28] "location_of_homicide"
## [29] "location_of_homicide_specify"
## [30] "number_of_victims_killed"
## [31] "number_of_victims_injured"
## [32] "number_of_shooting_injuries"
## [33] "number_of_other_injuries"
## [34] "students_killed"
## [35] "staff_killed"
## [36] "students_injured"
## [37] "staff_injured"
## [38] "hostages"
## [39] "weapon_used"
## [40] "how_was_weapon_obtained"
## [41] "when_was_weapon_obtained"
## [42] "other_weapons"
## [43] "preplanned"
## [44] "degree_of_planning"
## [45] "inspired_by_or_studied_other_perpetrators"
## [46] "inspired_by_or_studied_other_perpetrators_specify"
## [47] "connected_to_another_homicide"
## [48] "connected_to_another_homicide_specify"
## [49] "active_shooter"
## [50] "gang_group_related"
## [51] "drive_by"
## [52] "domestic"
## [53] "situation"
## [54] "on_scene_outcome"
## [55] "conviction_outcome"
## [56] "short_summary"
## [57] "full_description"
## [58] "articles"
## [59] "school_type_group"
## [60] "school_type"
## [61] "endowment_in_millions"
## [62] "hbcu"table(all_shootings$school_type_group)##
## Higher Ed K-12
## 280 398glimpse(all_shootings)## Rows: 678
## Columns: 62
## $ id_number <dbl> 1, 2, 3, 4, 5, 6, 7,…
## $ full_date <chr> "01/19/2000", "02/29…
## $ month <dbl> 1, 2, 3, 5, 5, 5, 7,…
## $ day <dbl> 19, 29, 10, 5, 10, 2…
## $ year <dbl> 2000, 2000, 2000, 20…
## $ day_of_week <chr> "Wednesday", "Tuesda…
## $ time_of_day <chr> "Afternoon", "Mornin…
## $ time_homicide_occurred <time> 14:13:00, 10:00:00,…
## $ during_class <chr> "No evidence of", "Y…
## $ school_name <chr> "Ridgewood High Scho…
## $ address <chr> "7650 Orchid Lake Rd…
## $ city <chr> "New Port Richey", "…
## $ state <chr> "FL", "MI", "GA", "C…
## $ zip_code <chr> "34653", "48505", "3…
## $ density <chr> "Suburban", "Urban",…
## $ latitude <dbl> 28.27111, 43.00643, …
## $ longitude <dbl> -82.68698, -83.66535…
## $ industry_type_of_business <chr> "Educational Institu…
## $ school_level <chr> "High", "Elementary"…
## $ number_of_employees_at_site <chr> "51-150", "Unknown",…
## $ enrollment <dbl> 1604, NA, 904, 1412,…
## $ accessibility_to_public <chr> "Publicly Accessible…
## $ gun_free_zone <chr> "Yes", "Yes", "Yes",…
## $ security_systems <chr> "School deputy on si…
## $ lockdown <chr> "No evidence of", "N…
## $ how_perpetrator_accessed <chr> "No Access Restricti…
## $ armed_person_on_scene <chr> "Yes", "No evidence …
## $ location_of_homicide <chr> "Outside", "Inside",…
## $ location_of_homicide_specify <chr> "Parking Lot", "Comm…
## $ number_of_victims_killed <dbl> 1, 1, 2, 1, 1, 1, 2,…
## $ number_of_victims_injured <dbl> 0, 0, 1, 1, 0, 0, 0,…
## $ number_of_shooting_injuries <dbl> 0, 0, 1, 0, 0, 0, 0,…
## $ number_of_other_injuries <dbl> 0, 0, 0, 1, 0, 0, 0,…
## $ students_killed <chr> "Yes", "Yes", "Yes",…
## $ staff_killed <chr> "No evidence of", "N…
## $ students_injured <chr> "No evidence of", "N…
## $ staff_injured <chr> "No evidence of", "N…
## $ hostages <chr> "No evidence of", "N…
## $ weapon_used <chr> "Handgun", "Handgun"…
## $ how_was_weapon_obtained <chr> "Stolen", "Stolen", …
## $ when_was_weapon_obtained <chr> "0 - 6 months prior"…
## $ other_weapons <chr> NA, "Knife", NA, NA,…
## $ preplanned <chr> "No evidence of", "N…
## $ degree_of_planning <chr> NA, NA, NA, NA, NA, …
## $ inspired_by_or_studied_other_perpetrators <chr> "No evidence of", "N…
## $ inspired_by_or_studied_other_perpetrators_specify <chr> NA, NA, NA, NA, NA, …
## $ connected_to_another_homicide <chr> "No evidence of", "N…
## $ connected_to_another_homicide_specify <chr> NA, NA, NA, NA, NA, …
## $ active_shooter <chr> "No evidence of", "N…
## $ gang_group_related <chr> "No evidence of", "N…
## $ drive_by <chr> "No evidence of", "N…
## $ domestic <chr> "No evidence of", "N…
## $ situation <chr> "Accident/Negligence…
## $ on_scene_outcome <chr> "Arrested on Scene",…
## $ conviction_outcome <chr> "Manslaughter by Cul…
## $ short_summary <chr> "Student shot throug…
## $ full_description <chr> "On Wednesday, Janua…
## $ articles <chr> "https://www.tampaba…
## $ school_type_group <chr> "K-12", "K-12", "K-1…
## $ school_type <chr> NA, NA, NA, NA, NA, …
## $ endowment_in_millions <chr> NA, NA, NA, NA, NA, …
## $ hbcu <chr> NA, NA, NA, NA, NA, …class(all_shootings$number_of_victims_killed)## [1] "numeric"summary(all_shootings$number_of_victims_killed)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.000 1.000 1.000 1.326 1.000 32.000table(all_shootings$number_of_victims_killed)##
## 1 2 3 4 5 6 9 10 17 21 26 32
## 604 44 14 5 3 1 2 1 1 1 1 1 # Select some variables of interest and see if there is any obvious correlations using the COR command
cor(
all_shootings$enrollment,
all_shootings$number_of_victims_killed,
use = "complete.obs"
)## [1] 0.03349962# Near 1 means a strong relationship, and near 0 means no relationshipThis information shows us that there is not a strong correlation between the size of the school and the number of fatalities. Essentially, larger schools typically do not experience more fatalities than smaller schools.
table(all_shootings$day_of_week)##
## Friday Monday Saturday Sunday Thursday Tuesday Wednesday
## 129 100 71 73 102 98 105# one of the scholarly aricles I was reading says that shootings have gotten more deadly over time. Let's run a correlation to see if that may be true
# A positive result indicates higher fatalities per incident over time; a negative result indicates the opposite.
cor(
all_shootings$year,
all_shootings$number_of_victims_killed,
use = "complete.obs"
)## [1] -0.01116037The negative result above shows us that the number of fatalities at each incident have decreased over time.
# Examine the same variables visually using the PAIRS command
pairs(all_shootings[, c("enrollment", "year", "number_of_victims_killed")])
# Select two variables that seem correlated (positively or negatively) and examine them using PEARSON, SPEARMAN or KENDALL (depending on which is more appropriate)
#I want to measure the relationship between the number of victims injured and the number of victims killed
names(all_shootings)[grepl("injur", names(all_shootings))]## [1] "number_of_victims_injured" "number_of_shooting_injuries"
## [3] "number_of_other_injuries" "students_injured"
## [5] "staff_injured"# first, we have to check and see if these values are considered numbers/numeric so we can compare them
class(all_shootings$number_of_victims_injured)## [1] "numeric"cor.test(all_shootings$number_of_victims_injured, all_shootings$number_of_victims_killed, method = "spearman", exact = FALSE)##
## Spearman's rank correlation rho
##
## data: all_shootings$number_of_victims_injured and all_shootings$number_of_victims_killed
## S = 34663260, p-value < 2.2e-16
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
## rho
## 0.3326825Spearman’s correlation is 0.33 which shows us that school homicide incidents involving more injuries tend to also involve more fatalities.
I used Spearman’s correlation because my data is very heavily skewed since most events only had one death and a few involved significantly more fatalities. The results show a small positive correlation (ρ = 0.333, p < .001) which communicates to me that incidents with more injuries tended to have more fatalities.