DACSS 697 Final Project

The data, research process, methodology, results, and takeaways from my DACSS 697: Text as Data final project

Title and Author:

“How Firearm Legislation Varies Across States for those with Domestic Violence-Related Records - and how it Relates to Firearm Homicide Rates”

Megan Georges – DACSS 697: Text as Data - Spring 2022

Abstract

Recent research finds that 50-60% of female homicides are perpetrated by an intimate partner, with over half of those executed using a firearm. Additionally, over 4.5 million U.S. women have reported being threatened with a firearm by an intimate partner. In 1996, federal law banned firearm possession for individuals convicted of a felony or domestic violence (DV) misdemeanor or who have a DV-related protective order against them. States have since implemented DV firearm provisions that establish additional regulations on the matter, which this research uses Structural Topic Modeling to measure. The research examines state legislative differences in topic prevalence for explicate prohibition of purchase/possession, expanded protective order eligibility, and firearm relinquishment/seizure requirements against intimate partner firearm homicide (IPFH) rates. Preliminary findings suggest that states that fail to explicitly ban DV firearm possession experience higher IPFH rates. The research establishes grounds for continued research that could invoke nation-wide policy initiatives.

Research Question:

Is there a relationship between the restrictiveness of a state’s firearm regulations/laws for individuals with domestic violence related records and homicide rates perpetrated using a firearm?

Background:

Since 1996, Federal law has banned the possession of firearms for individuals convicted of a felony or a domestic violence (DV) misdemeanor, or who have a DV-related protective order against them (U.S. Department of Justice, 2020). However, states vary significantly in the number of and restrictiveness of firearms laws, particularly as they pertain to DV-offenders. For example, some states fail to explicitly ban the purchase or possession of firearms for DV perpetrators. While federal law takes priority, the explicit statement of firearm restrictions in state laws often is followed by additional provisions specifying the definition of an intimate partner, what types of DV offenses/circumstances are included, how firearms should be relinquished or seized from offenders, and what the consequences are for violating provisions (Giffords Law Center, 2021).

Recent research finds that 50-60% of female homicides are perpetrated by a current or former intimate partner, and over half of those are executed using a firearm (Disarm Domestic Violence, 2022). Giffords Law Center (2021) additionally finds that 4.5 million U.S. women have reported being threatened with a firearm by an intimate partner. Researchers believe that the issue is even more prevalent than the data indicate due to stigmatization and the complexity of reporting and tracking intimate partner violence.

However, due to increased awareness of intimate partner violence (IPV) and its frequent association to firearm violence, many states have implemented legal provisions to prevent perpetrators from possessing firearms. States have implemented legislation that specifies who may petition for a DV protective order and the scenarios that would result in firearm prohibitions for the abuser or alleged abuser. Additional provisions pertain to an offender’s duty to relinquish firearms and detail the process law enforcement may follow to remove firearms from those with a DV-related criminal or civil imposition (Disarm Domestic Violence, 2022).

Description of Data

Center for Disease Control (CDC) - CDC WONDER

The Centers for Disease Control and Prevention (CDC) provides CDC Wonder, which is an online database that allows for the collection and analysis of public health data. I am using the Underlying Cause of Death database, which allows the selection of ‘Homicide’ as the ‘Injury Intent’ and the specification of ‘Firearm’ as the ‘Injury Mechanism’. This information is determined through death certificates.

I decided to use an average across 3 years to account for any single year abnormalities. The information on state firearm laws (data below - ‘Disarm Domestic Violence’) was updated mid-2021 according to the source. I do not have information on how long each state has had each firearm provision enacted, so I am using homicide rate averages across 2017 to 2019 (also excluding 2020 due to onset of COVID-19 as potential cause of irregularity). I’ve also included overall rates and then by gender. Since previous research indicates that a high number of female homicides by firearm has been perpetrated by an intimate partner (over half, as discussed in Background section), I will evaluate female homicide rates in relation to state firearm legislation. Due to the lack of data on confirmed IPH in many states, overall firearm homicides is the closest to that information. However, included below is a source that provides IPH data for some states.

Source: https://wonder.cdc.gov/ucd-icd10.html

# Overall Homicide Rates
HomRates <- suppressMessages(read.delim("../../DACSS 697/DV Research/Homicide_Firearm_2017through2019.txt"))
HomRates <- select(HomRates, State, Deaths, Population, Crude.Rate) %>% mutate(across(everything(), ~ifelse(.=="", NA, as.character(.)))) %>% na.omit()
HomRates <- filter(HomRates, !str_detect(State, "District of Columbia"))
HomRates$Deaths <- as.numeric(HomRates$Deaths)
HomRates$Population <- as.numeric(HomRates$Population)
HomRates$Crude.Rate <- as.numeric(HomRates$Crude.Rate)
A <- function(x) x/3
HomRates$Deaths <- sapply(HomRates$Deaths, A)
HomRates$Population <- sapply(HomRates$Population, A)
HomRates <- HomRates %>%
  mutate_at(vars(Population, Deaths), funs(round(., 0)))

# Homicide Rates by Gender
HomRatesG <- suppressMessages(read.delim("../../DACSS 697/DV Research/CDC_Gun_Hom_Gender.txt"))
HomRatesG <- select(HomRatesG, State, Gender, Deaths, Crude.Rate) %>%
  rename(States = State)
HomRatesG <- filter(HomRatesG, !str_detect(States, "District of Columbia")) 
HomRatesG$Deaths <- as.numeric(HomRatesG$Deaths)
HomRatesG$Crude.Rate <- as.numeric(HomRatesG$Crude.Rate)
A <- function(x) x/3
HomRatesG$Deaths <- sapply(HomRatesG$Deaths, A)
HomRatesG <- HomRatesG %>%
  mutate_at(vars(Deaths), funs(round(., 0)))
HomRatesG <- HomRatesG %>% 
  mutate(across(everything(), ~ifelse(.=="", NA, as.character(.)))) %>%
  filter(!str_detect(States, "NA"))
HomRatesG <- HomRatesG %>%
  pivot_wider(names_from = Gender, values_from = c(Deaths, Crude.Rate))  

# Combine overall rates and gender rates to one dataframe
CDCHomRates <- cbind.data.frame(HomRates, HomRatesG)
CDCHomRates <- select(CDCHomRates, -c(States))
CDCHomRates <- CDCHomRates[, c("State", "Crude.Rate_Female", "Crude.Rate_Male", "Crude.Rate", "Deaths_Female", "Deaths_Male", "Deaths", "Population")]
CDCHomRates <- rename_(CDCHomRates, "Crude_Total" = "Crude.Rate", "Deaths_Total" = "Deaths")
# Fixing suppressed and unreliable rates
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "Maine"] <- "0.7"
CDCHomRates$Crude.Rate_Male[CDCHomRates$State == "Maine"] <- "0.9"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "Vermont"] <- "1.1"
CDCHomRates$Crude.Rate_Male[CDCHomRates$State == "Vermont"] <- "1.3"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "Wyoming"] <- "1.5"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "New Hampshire"] <- "0.8"

CDCHomRates$Deaths_Female[CDCHomRates$State == "Hawaii"] <- "3"
CDCHomRates$Deaths_Female[CDCHomRates$State == "North Dakota"] <- "2"
CDCHomRates$Deaths_Female[CDCHomRates$State == "South Dakota"] <- "1"
CDCHomRates$Deaths_Female[CDCHomRates$State == "Rhode Island"] <- "3"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "Hawaii"] <- "0.1"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "North Dakota"] <- "0.2"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "South Dakota"] <- "0.1"
CDCHomRates$Crude.Rate_Female[CDCHomRates$State == "Rhode Island"] <- "0.2"

# Present data with table
kable(CDCHomRates, col.names = c("State", "Female", "Male", "Total", "Female", "Male", "Total", "Population"), 
      align = c('c', 'c', 'c', 'c', 'c', 'c', 'c', 'c')) %>%
  add_header_above(c("", "Crude Rate (per 100,000)"=3, "Number of Deaths"=3, ""))%>%
  add_header_above(c("Homicide by Firearm (Averages 2017-2019)"=8)) %>%
    kable_styling(fixed_thead = TRUE)%>%
  scroll_box(width = "100%", height = "600px") %>%
  footnote(general = "The CDC marks rates as unreliable when death counts are fewer than 20, which applies here to Female Crude Rates for Hawaii, North Dakota, Rhode Island, and South Dakota")

Homicide by Firearm (Averages 2017-2019)
	Crude Rate (per 100,000)			Number of Deaths
State	Female	Male	Total	Female	Male	Total	Population
Alabama	3.3	17.1	10.0	82	404	487	4888601
Alaska	3.3	8.6	6.1	12	33	45	736259
Arizona	1.4	6.8	4.1	52	242	294	7155544
Arkansas	2.5	10.8	6.6	38	160	198	3011969
California	1	5.7	3.3	195	1124	1319	39535307
Colorado	1	5	3.0	28	142	171	5687151
Connecticut	0.5	3.2	1.8	9	55	64	3575379
Delaware	1.7	8.8	5.1	8	41	49	967625
Florida	1.6	7.8	4.6	172	807	979	21253821
Georgia	1.9	11	6.3	105	562	667	10522092
Hawaii	0.1	1.4	0.9	3	10	13	1421300
Idaho	1	1.9	1.4	9	17	25	1752739
Illinois	1.4	11.9	6.6	90	746	836	12738308
Indiana	1.8	9.1	5.4	62	302	363	6696972
Iowa	0.5	2.7	1.6	8	42	50	3152309
Kansas	1.5	6.7	4.1	22	98	120	2912647
Kentucky	1.8	7.8	4.7	40	172	212	4463421
Louisiana	3	19.9	11.2	72	453	524	4664368
Maine	0.7	0.9	0.8	5	6	11	1339508
Maryland	1.4	13.9	7.4	43	406	449	6046858
Massachusetts	0.3	2.7	1.5	11	90	101	6884824
Michigan	1.5	7.8	4.6	74	384	458	9981694
Minnesota	0.4	2.5	1.5	13	69	82	5609139
Mississippi	3.6	18.8	11.0	55	272	327	2982260
Missouri	3.1	15.1	9.0	97	455	552	6125804
Montana	1.5	2.6	2.0	8	14	22	1060525
Nebraska	0.9	2.4	1.6	9	23	31	1927917
Nevada	2.1	7.3	4.7	32	111	143	3037529
New Hampshire	0.8	1.2	1.0	6	8	14	1352988
New Jersey	0.6	4.5	2.5	26	198	224	8932118
New Mexico	2.5	10.2	6.3	26	105	132	2093442
New York	0.4	3	1.7	42	289	331	19615056
North Carolina	1.5	8.7	5.0	78	439	517	10381708
North Dakota	0.2	2.1	1.4	2	8	10	759177
Ohio	1.6	8.4	4.9	97	481	578	11679050
Oklahoma	2	9.2	5.6	41	179	220	3943638
Oregon	0.6	2.8	1.7	14	59	73	4183742
Pennsylvania	1.3	7.9	4.5	83	494	577	12804862
Rhode Island	0.2	1.8	1.1	3	9	12	1058772
South Carolina	2.4	13.2	7.6	63	325	388	5085737
South Dakota	0.1	2.7	1.4	1	12	13	878853
Tennessee	2.3	12.2	7.1	80	402	482	6771723
Texas	1.5	7.1	4.3	216	1017	1233	28667441
Utah	0.8	2.1	1.5	12	34	46	3156299
Vermont	1.1	1.3	1.2	3	4	7	624648
Virginia	1.4	6.7	4.0	62	279	340	8507741
Washington	1	3.5	2.2	36	132	168	7518742
West Virginia	2	5.8	3.9	19	52	70	1804612
Wisconsin	1.1	4.2	2.7	33	122	155	5810495
Wyoming	1.5	2.6	2.1	4	8	12	578604

Note: The CDC marks rates as unreliable when death counts are fewer than 20, which applies here to Female Crude Rates for Hawaii, North Dakota, Rhode Island, and South Dakota

National Violent Death Reporting System (NVDRS) - Web-Based Injury Statistics Query and Reporting System (WISQARS)

The National Violent Death Reporting System (NVDRS) is a product of the CDC and specifically the Web-based Injury Statistics Query and Reporting System (WISQARS). This database provides data specifically on confirmed homicides perpetrated by an intimate partner and using a firearm, as opposed to the CDC WONDER dataset above of all firearm homicides. Confirmed intimate partner homicide (IPH) rates are not available for all states or years, but that will likely change as the CDC began providing NVDRS funding to all states in 2018. The system allows for states to combine law enforcement reports, medical examiner/coroner reports, and death certificates when evaluating and reporting public health matters like homicide.

For the states included in the NVDRS between 2017 and 2019, I will analyze crude rates in relation to state firearm laws.

Source: https://www.cdc.gov/injury/wisqars/nvdrs/

# Confirmed intimate partner homicides (not all states are included)
NVDRSdata <- suppressMessages(read_csv("../../DACSS 697/DV Research/NVDRSdf.csv"))

NVDRSdata <- NVDRSdata %>%
  select(State, Average_Deaths, Crude_Rate)%>%
  filter(!str_detect(`State`, "TOTAL"))

# Present data with table
kable(NVDRSdata, col.names = c("State", "Number of Deaths", "Crude Rate")) %>%
  kable_styling() %>%
  add_header_above(c("Confirmed Intimate Partner Homicides Using a Firearm (2017-2019 Averages)"=3)) %>%
  scroll_box(width = "100%", height = "600px") %>%
  footnote(general = "if NA, value has been suppressed by the CDC as a privacy constraint:",
  number = c("number of deaths count is between 0 and 9", 
    "crude rate unable to be calculated due to suppressed death count"))

Confirmed Intimate Partner Homicides Using a Firearm (2017-2019 Averages)
State	Number of Deaths	Crude Rate
Alaska	14	0.63
Arizona	84	0.39
California	117	0.21
Colorado	47	0.28
Connecticut	15	0.14
Delaware	10	0.34
Georgia	149	0.47
Illinois	68	0.21
Indiana	57	0.28
Iowa	12	0.13
Kansas	35	0.4
Kentucky	70	0.52
Maine	13	0.32
Maryland	33	0.18
Massachusetts	19	0.09
Michigan	109	0.36
Minnesota	22	0.13
Nevada	50	0.55
New Hampshire	10	0.25
New Jersey	27	0.1
New Mexico	39	0.62
North Carolina	135	0.43
Ohio	133	0.38
Oklahoma	88	0.74
Oregon	35	0.28
Pennsylvania	92	0.29
Rhode Island	Suppressed	Suppressed
South Carolina	88	0.58
Utah	12	0.13
Vermont	Suppressed	Suppressed
Virginia	98	0.38
Washington	59	0.27
West Virginia	37	0.68
Wisconsin	52	0.3

Note: if NA, value has been suppressed by the CDC as a privacy constraint: ¹ number of deaths count is between 0 and 9 ² crude rate unable to be calculated due to suppressed death count

Disarm Domestic Violence

Disarm Domestic Violence is a website that compiles information on each state’s domestic violence related legislation as it pertains to firearms. Key focal points of the source are prohibition, specification of the victim-perpetrator relationship, protective orders, judicial authority, the firearm removal process, and punishments for violations.

Source: https://www.disarmdv.org/#

# Read in the data using web-scraping

URL <- "https://www.disarmdv.org/state/"

State <- c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming')

URLS <- URL

# loop through each state-specific URL 
for (i in 1:length(State)){
  URLS <- c(URLS, paste("https://www.disarmdv.org/state/", State[i], sep = ""))
}
StateURL <- paste(URLS, "/?sec=law", sep="")
StateURL <- StateURL[2:51]
disarmdv <- c()
css_selector <- ".auto-navigation-content"

for (i in 1:length(StateURL)){
  
  laws <- StateURL[i] %>% 
  read_html() %>%
  html_nodes(css = css_selector) %>%
  html_text()

  
  disarmdv <- c(disarmdv, laws)
}

Methodology

With this research, I am trying to identify the presence or omission of certain firearm provisions for each state, as well as how prevalent certain topics are in a state’s legislation. The data source for this information covers the same topics across all states - and first identifies whether certain measures are explicitly defined by the state then elaborates on the measures for the states that include said measure. Therefore, I will employ Structural Topic Modeling (STM) to allow for the identification of topics across documents and the measurement of prevalence of key topics for each state. STM is particularly useful because it allows for the evaluation of covariates, which in this case are differing states and homicide rates.

Pre-processing and Representing Texts

First, I will perform pretty extensive pre-processing. Based upon running some initial topic modeling, the documents of the corpus share a lot of similarities of text because the source includes sectioned headlines about each firearm provision for each state. I will need to remove the headlines from each document to allow the topic modeling to focus on the actual content within each section.

For example, the first headline in each document reads “DOMESTIC VIOLENCE FIREARM PROHIBITIONS”. However, the text will follow with “Georgia does not prohibit persons convicted of misdemeanor crimes of domestic violence from purchasing or possessing firearms or ammunition.” or “California prohibits the following from purchasing or possessing firearms or ammunition…” and then includes elaboration of these prohibitions.

Also, there are a lot of phrases or words that provide little insight when separated, so I will compound certain words or phrases to help the topic model better identify topics by maintaining context. I used quanteda’s keyword-in-contexts kwic() function to detect these phrases by inputing a keyword like ‘prohibit’ and identifying which words are present across the corpus that provide key context to the word.

# create data frame from web-scraped data

disarmData <- as.data.frame(disarmdv)
disarmData$state <- c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming')
disarmData <- disarmData[, c("state", "disarmdv")]
disarmData <- rename_(disarmData, "Text" = "disarmdv")

# give preview of how text scraped in from website for 1st state - Alabama
kable(head(disarmData, 1)) %>% kable_styling()

state

Text

alabama

Alabama Law ALABAMA DOMESTIC VIOLENCE FIREARM PROHIBITIONS Alabama Domestic Violence Firearm Purchase and Possession Prohibitions Alabama prohibits the following individuals from owning a firearm, possessing a firearm, or having a firearm in their control: Persons convicted of a misdemeanor offense of domestic violence; and Persons subject to a valid protection order for domestic abuse.1 “Valid protection order” is defined as “an order issued after a hearing of which the person received actual notice, and at which the person had an opportunity to participate, that does any of the following: Restrains the person from harassing, stalking, or threatening a  qualified individual* or child of the qualified individual or person or engaging in other conduct that would place a qualified individual in reasonable fear of bodily injury to the individual or child and that includes a finding that the person represents a credible threat to the physical safety of the qualified individual or child. By its terms, explicitly prohibits the use, attempted use, or threatened use of physical force against the qualified individual or child that would reasonably be expected to cause bodily injury.”2 ALABAMA CIVIL PROTECTION ORDER FIREARM REMOVAL Domestic Violence Civil Protection Orders That Require Firearm Removal Alabama law does not require the removal of firearms from persons subject to domestic violence protection orders. Alabama law does allow a judge issuing an ex parte protection order, an ex parte modification of a protection order, a final protection order, or a modification of a protection order issued after notice and hearing to “[o]rder other relief as it deems necessary to provide for the safety and welfare of the plaintiff or any children and any person designated by the court.”3 Individuals Who May Petition for a Protection Order The following persons may petition for a protection order: A spouse (including a common law spouse); A former spouse (including a common law former spouse); A person with whom the defendant has a child in common, regardless of whether the victim or defendant have ever been married and regardless of whether they are currently residing or have in the past resided together in the same household; A person who has or had a dating relationship with the defendant; A person who is or was cohabiting with the defendant and who is in, or was engaged in, a romantic or sexual relationship with the defendant; A relative of a person defined in (e) who also lived with the defendant; or An individual who is a parent, stepparent, child, or stepchild and who is in or has maintained a living arrangement with the defendant.4 Penalties for Violation A violation of a protective order is a Class A misdemeanor.5

# merge all datasets together to create metadata 
MetaData <- cbind(CDCHomRates, disarmData)
MetaData <- select(MetaData, "State", "Crude.Rate_Female", "Crude.Rate_Male", "Crude_Total", "Deaths_Female", "Deaths_Male", "Deaths_Total", "Population", "Text" )

# Identify and remove words in all caps because those are the headlines
# consistent across each document
MetaData$Text <- unlist(str_remove_all(MetaData$Text, "\\b[A-Z]+\\b"))

# Create corpus and set document names 

Meta_corpus <- corpus(MetaData, text_field = 'Text')
docnames(Meta_corpus) <- c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming')

summary(Meta_corpus)

Corpus consisting of 50 documents, showing 50 documents:

           Text Types Tokens Sentences          State
        alabama   197    479         5        Alabama
         alaska   176    467         8         Alaska
        arizona   180    490         7        Arizona
       arkansas   127    219         5       Arkansas
     california   485   1716        17     California
       colorado   252    996         4       Colorado
    connecticut   308   1264         3    Connecticut
       delaware   423   1506        11       Delaware
        florida   264    779         7        Florida
        georgia   116    213         4        Georgia
         hawaii   229    608         5         Hawaii
          idaho   146    320         5          Idaho
       illinois   369   1349         7       Illinois
        indiana   174    459         6        Indiana
           iowa   224    661         6           Iowa
         kansas   179    397         3         Kansas
       kentucky   242    610         7       Kentucky
      louisiana   473   2077        20      Louisiana
          maine   378   1403         6          Maine
       maryland   249    976        10       Maryland
  massachusetts   347   1303         7  Massachusetts
       michigan   259    697         5       Michigan
      minnesota   426   1682         4      Minnesota
    mississippi   177    388         3    Mississippi
       missouri   152    321         5       Missouri
        montana   187    517         4        Montana
       nebraska   161    343         3       Nebraska
         nevada   242    956         5         Nevada
  new-hampshire   366   1326        23  New Hampshire
     new-jersey   259   1145         9     New Jersey
     new-mexico   368   1298        10     New Mexico
       new-york   431   1724         9       New York
 north-carolina   405   1623        23 North Carolina
   north-dakota   254    626         4   North Dakota
           ohio   289    836         9           Ohio
       oklahoma   255    764         7       Oklahoma
         oregon   232    664         5         Oregon
   pennsylvania   612   3515        49   Pennsylvania
   rhode-island   373   1478         8   Rhode Island
 south-carolina   163    453         2 South Carolina
   south-dakota   209    592         6   South Dakota
      tennessee   249    751         3      Tennessee
          texas   224    646         4          Texas
           utah   244    817         2           Utah
        vermont   364   1204         9        Vermont
       virginia   268    863         5       Virginia
     washington   494   2361        14     Washington
  west-virginia   330   1311        11  West Virginia
      wisconsin   443   2103        14      Wisconsin
        wyoming   169    383         6        Wyoming
 Crude.Rate_Female Crude.Rate_Male Crude_Total Deaths_Female
               3.3            17.1        10.0            82
               3.3             8.6         6.1            12
               1.4             6.8         4.1            52
               2.5            10.8         6.6            38
                 1             5.7         3.3           195
                 1               5         3.0            28
               0.5             3.2         1.8             9
               1.7             8.8         5.1             8
               1.6             7.8         4.6           172
               1.9              11         6.3           105
               0.1             1.4         0.9             3
                 1             1.9         1.4             9
               1.4            11.9         6.6            90
               1.8             9.1         5.4            62
               0.5             2.7         1.6             8
               1.5             6.7         4.1            22
               1.8             7.8         4.7            40
                 3            19.9        11.2            72
               0.7             0.9         0.8             5
               1.4            13.9         7.4            43
               0.3             2.7         1.5            11
               1.5             7.8         4.6            74
               0.4             2.5         1.5            13
               3.6            18.8        11.0            55
               3.1            15.1         9.0            97
               1.5             2.6         2.0             8
               0.9             2.4         1.6             9
               2.1             7.3         4.7            32
               0.8             1.2         1.0             6
               0.6             4.5         2.5            26
               2.5            10.2         6.3            26
               0.4               3         1.7            42
               1.5             8.7         5.0            78
               0.2             2.1         1.4             2
               1.6             8.4         4.9            97
                 2             9.2         5.6            41
               0.6             2.8         1.7            14
               1.3             7.9         4.5            83
               0.2             1.8         1.1             3
               2.4            13.2         7.6            63
               0.1             2.7         1.4             1
               2.3            12.2         7.1            80
               1.5             7.1         4.3           216
               0.8             2.1         1.5            12
               1.1             1.3         1.2             3
               1.4             6.7         4.0            62
                 1             3.5         2.2            36
                 2             5.8         3.9            19
               1.1             4.2         2.7            33
               1.5             2.6         2.1             4
 Deaths_Male Deaths_Total Population
         404          487    4888601
          33           45     736259
         242          294    7155544
         160          198    3011969
        1124         1319   39535307
         142          171    5687151
          55           64    3575379
          41           49     967625
         807          979   21253821
         562          667   10522092
          10           13    1421300
          17           25    1752739
         746          836   12738308
         302          363    6696972
          42           50    3152309
          98          120    2912647
         172          212    4463421
         453          524    4664368
           6           11    1339508
         406          449    6046858
          90          101    6884824
         384          458    9981694
          69           82    5609139
         272          327    2982260
         455          552    6125804
          14           22    1060525
          23           31    1927917
         111          143    3037529
           8           14    1352988
         198          224    8932118
         105          132    2093442
         289          331   19615056
         439          517   10381708
           8           10     759177
         481          578   11679050
         179          220    3943638
          59           73    4183742
         494          577   12804862
           9           12    1058772
         325          388    5085737
          12           13     878853
         402          482    6771723
        1017         1233   28667441
          34           46    3156299
           4            7     624648
         279          340    8507741
         132          168    7518742
          52           70    1804612
         122          155    5810495
           8           12     578604

# Use kwic() to identify context for key words - to create compounded tokens
KeyWords <- tokens(Meta_corpus)
kwic(KeyWords, pattern = "remov", valuetype = "regex", window = 2) %>%
  kable() %>% kable_styling() %>% scroll_box(width = "100%", height = "400px")

docname	from	to	pre	keyword	post	pattern
alabama	208	208	Require Firearm	Removal	Alabama law	remov
alabama	215	215	require the	removal	of firearms	remov
alaska	110	110	Require Firearm	Removal	Alaska law	remov
alaska	117	117	require the	removal	of firearms	remov
alaska	358	358	] 3	Removal	Process court	remov
arizona	162	162	Require Firearm	Removal	If a	remov
arizona	213	213	for the	removal	of firearms	remov
arizona	396	396	. 5	Removal	Process defendant	remov
arkansas	53	53	Require Firearm	Removal	Arkansas does	remov
arkansas	58	58	not require	removal	of firearms	remov
california	98	98	Require Firearm	Removal	Any protective	remov
california	128	128	require the	removal	of firearms	remov
california	475	475	. 7	Removal	Process Upon	remov
california	945	945	Exemption from	Removal	Requirement The	remov
colorado	100	100	that Require	Removal	Colorado requires	remov
colorado	293	293	include firearm	removal	only for	remov
colorado	327	327	. 8	Removal	Process court	remov
connecticut	545	545	that Require	Removal	Connecticut requires	remov
connecticut	548	548	Connecticut requires	removal	of firearms	remov
connecticut	703	703	. 12	Removal	Process person	remov
delaware	139	139	that Require	Removal	Delaware law	remov
delaware	148	148	firearms be	removed	from persons	remov
delaware	190	190	ammunition be	removed	from the	remov
delaware	394	394	” 6	Removal	Process judge	remov
florida	98	98	that Require	Removal	Florida law	remov
florida	105	105	require the	removal	of firearms	remov
florida	126	126	order the	removal	of firearms	remov
florida	607	607	Exemption from	Removal	Requirement Florida	remov
florida	625	625	protection from	removal	of firearms	remov
georgia	55	55	that Require	Removal	Georgia does	remov
georgia	60	60	not require	removal	of firearms	remov
hawaii	180	180	that Require	Removal	Hawaii requires	remov
hawaii	183	183	Hawaii requires	removal	of any	remov
hawaii	323	323	above .	Removal	Process Hawaii	remov
idaho	57	57	that Require	Removal	Idaho does	remov
idaho	62	62	not require	removal	of firearms	remov
illinois	463	463	that Require	Removal	If the	remov
illinois	866	866	” 11	Removal	Process If	remov
indiana	75	75	that Require	Removal	Indiana does	remov
indiana	80	80	not require	removal	of firearms	remov
indiana	288	288	. 5	Removal	Process As	remov
iowa	83	83	that Require	Removal	court that	remov
iowa	302	302	” 5	Removal	Process court	remov
iowa	475	475	Exemption from	Removal	Requirement The	remov
kansas	215	215	that Require	Removal	Kansas does	remov
kansas	220	220	not require	removal	of firearms	remov
kentucky	120	120	that Require	Removal	Kentucky does	remov
kentucky	125	125	not require	removal	of firearms	remov
louisiana	221	221	that Require	Removal	Upon the	remov
louisiana	807	807	” 12	Removal	Process Upon	remov
maine	274	274	that Require	Removal	court issuing	remov
maine	867	867	by :	Removing	that person	remov
maine	1022	1022	” 11	Removal	Process If	remov
maryland	318	318	Require Firearm	Removal	If a	remov
maryland	667	667	. 9	Removal	Process respondent	remov
massachusetts	343	343	that Require	Removal	court issuing	remov
massachusetts	633	633	. 10	Removal	Process If	remov
massachusetts	1136	1136	and licenses	removed	pursuant to	remov
massachusetts	1196	1196	and licenses	removed	pursuant to	remov
michigan	78	78	that Require	Removal	Michigan does	remov
michigan	83	83	not require	removal	of firearms	remov
minnesota	352	352	that Require	Removal	court issuing	remov
minnesota	373	373	order the	removal	of any	remov
minnesota	713	713	. 11	Removal	Process court	remov
minnesota	1326	1326	of the	removal	of firearms	remov
mississippi	59	59	that Require	Removal	Mississippi does	remov
mississippi	64	64	not require	removal	of firearms	remov
missouri	59	59	that Require	Removal	Missouri does	remov
missouri	64	64	not require	removal	of firearms	remov
montana	156	156	that Require	Removal	Montana does	remov
montana	161	161	not require	removal	of firearms	remov
nebraska	104	104	that Require	Removal	Nebraska does	remov
nebraska	109	109	not require	removal	of firearms	remov
nevada	137	137	Require Firearm	Removal	Nevada does	remov
nevada	142	142	not require	removal	of firearms	remov
nevada	335	335	. 7	Removal	Process As	remov
new-hampshire	131	131	that Require	Removal	judge issuing	remov
new-hampshire	142	142	shall require	removal	of any	remov
new-hampshire	163	163	require such	removal	when issuing	remov
new-hampshire	432	432	” 7	Removal	Process judge	remov
new-hampshire	704	704	relinquished or	removed	pursuant to	remov
new-hampshire	791	791	relinquished or	removed	pursuant to	remov
new-jersey	314	314	that Require	Removal	judge issuing	remov
new-jersey	327	327	order the	removal	of firearms	remov
new-jersey	359	359	require the	removal	of firearms	remov
new-jersey	532	532	” 10	Removal	Process judge	remov
new-jersey	954	954	from the	Removal	Requirement The	remov
new-jersey	1030	1030	surrender or	remove	a firearm	remov
new-mexico	165	165	that Require	Removal	If a	remov
new-mexico	295	295	not require	removal	of firearms	remov
new-mexico	516	516	. 7	Removal	Process If	remov
new-york	256	256	that Require	Removal	court issuing	remov
new-york	968	968	. 11	Removal	Process Where	remov
north-carolina	183	183	that Require	Removal	Upon the	remov
north-carolina	695	695	” 7	Removal	Process Upon	remov
north-carolina	952	952	from the	Removal	Requirement Law	remov
north-dakota	202	202	Require Firearm	Removal	North Dakota	remov
north-dakota	210	210	require the	removal	of firearms	remov
north-dakota	415	415	. 4	Removal	Process If	remov
ohio	100	100	that Require	Removal	Ohio does	remov
ohio	106	106	explicitly require	removal	of firearms	remov
oklahoma	126	126	that Require	Removal	Oklahoma does	remov
oklahoma	131	131	not require	removal	of firearms	remov
oregon	325	325	that Require	Removal	Oregon does	remov
oregon	330	330	not require	removal	of firearms	remov
pennsylvania	167	167	that Require	Removal	court issuing	remov
pennsylvania	724	724	” 7	Removal	Process The	remov
rhode-island	116	116	that Require	Removal	person suffering	remov
rhode-island	588	588	. 8	Removal	Process judge	remov
south-carolina	433	433	that Require	Removal	South Carolina	remov
south-carolina	439	439	not require	removal	of firearms	remov
south-dakota	64	64	Require Firearm	Removal	South Dakota	remov
south-dakota	71	71	require the	removal	of firearms	remov
tennessee	96	96	that Require	Removal	The respondent	remov
tennessee	396	396	” 6	Removal	Process An	remov
texas	331	331	that Require	Removal	Texas law	remov
texas	337	337	not require	removal	of firearms	remov
utah	324	324	that Require	Removal	Utah law	remov
utah	330	330	not require	removal	of firearms	remov
vermont	258	258	that Require	Removal	Vermont does	remov
vermont	498	498	” 8	Removal	Process respondent	remov
virginia	244	244	that Require	Removal	The respondent	remov
washington	296	296	that Require	Removal	” Any	remov
washington	1126	1126	. 12	Removal	Process court	remov
washington	1135	1135	includes a	removal	requirement ”	remov
west-virginia	319	319	Require Firearm	Removal	If the	remov
west-virginia	757	757	; 9	Removal	Process If	remov
wisconsin	119	119	that Require	Removal	person subject	remov
wisconsin	496	496	. 10	Removal	Process If	remov
wisconsin	1193	1193	Exemption from	Removal	Requirement If	remov
wyoming	59	59	that Require	Removal	Wyoming law	remov
wyoming	66	66	require the	removal	of firearms	remov

kwic(KeyWords, pattern = "prohib", valuetype = "regex", window = 2) %>%
  kable() %>% kable_styling() %>% scroll_box(width = "100%", height = "400px")

docname	from	to	pre	keyword	post	pattern
alabama	10	10	and Possession	Prohibitions	Alabama prohibits	prohib
alabama	12	12	Prohibitions Alabama	prohibits	the following	prohib
alabama	169	169	, explicitly	prohibits	the use	prohib
alaska	10	10	and Possession	Prohibitions	Alaska may	prohib
alaska	13	13	Alaska may	prohibit	the person	prohib
alaska	77	77	does not	prohibit	firearm purchase	prohib
arizona	10	10	and Possession	Prohibitions	Arizona prohibits	prohib
arizona	12	12	Prohibitions Arizona	prohibits	persons convicted	prohib
arizona	48	48	Arizona may	prohibit	the subject	prohib
arizona	95	95	Arizona may	prohibit	the subject	prohib
arizona	174	174	of protection	prohibits	the defendant	prohib
arkansas	10	10	and Possession	Prohibitions	Arkansas does	prohib
arkansas	14	14	does not	prohibit	the following	prohib
california	10	10	and Possession	Prohibitions	California prohibits	prohib
california	12	12	Prohibitions California	prohibits	the following	prohib
california	520	520	shall also	prohibit	the respondent	prohib
california	548	548	respondent is	prohibited	from owning	prohib
california	1349	1349	respondent is	prohibited	from possessing	prohib
colorado	10	10	and Possession	Prohibitions	Colorado prohibits	prohib
colorado	12	12	Prohibitions Colorado	prohibits	the following	prohib
colorado	578	578	is not	prohibited	from possessing	prohib
connecticut	10	10	and Possession	Prohibitions	Under Connecticut	prohib
connecticut	286	286	4 Connecticut	prohibits	possession of	prohib
connecticut	379	379	5 Connecticut	prohibits	possession of	prohib
connecticut	555	555	who are	prohibited	from possessing	prohib
connecticut	727	727	ammunition is	prohibited	shall ,	prohib
delaware	10	10	and Possession	Prohibitions	Delaware prohibits	prohib
delaware	12	12	Prohibitions Delaware	prohibits	the following	prohib
delaware	108	108	she is	prohibited	from receiving	prohib
delaware	745	745	Delaware law	prohibits	the respondent	prohib
delaware	1168	1168	not otherwise	prohibited	from purchasing	prohib
florida	10	10	and Possession	Prohibitions	Florida does	prohib
florida	14	14	does not	prohibit	persons convicted	prohib
florida	32	32	. Florida	prohibits	persons subject	prohib
florida	676	676	unless otherwise	prohibited	by the	prohib
florida	690	690	firearm possession	prohibition	. 10	prohib
georgia	10	10	and Possession	Prohibitions	Georgia does	prohib
georgia	14	14	does not	prohibit	persons convicted	prohib
georgia	34	34	does not	prohibit	persons subject	prohib
hawaii	10	10	and Possession	Prohibitions	Hawaii prohibits	prohib
hawaii	12	12	Prohibitions Hawaii	prohibits	ownership ,	prohib
hawaii	109	109	2 Hawaii	prohibits	ownership ,	prohib
idaho	10	10	and Possession	Prohibitions	Idaho does	prohib
idaho	14	14	does not	prohibit	persons convicted	prohib
idaho	34	34	does not	prohibit	persons subject	prohib
illinois	10	10	and Possession	Prohibitions	Under Illinois	prohib
illinois	137	137	or Is	prohibited	from acquiring	prohib
illinois	299	299	terms explicitly	prohibits	the use	prohib
illinois	339	339	protection may	prohibit	a respondent	prohib
illinois	484	484	of protection	prohibits	the respondent	prohib
illinois	888	888	of protection	prohibits	the respondent	prohib
indiana	10	10	and Possession	Prohibitions	Indiana prohibits	prohib
indiana	12	12	Prohibitions Indiana	prohibits	persons convicted	prohib
iowa	10	10	and Possession	Prohibitions	Iowa prohibits	prohib
iowa	12	12	Prohibitions Iowa	prohibits	persons convicted	prohib
iowa	44	44	1 Iowa	prohibits	persons subject	prohib
iowa	59	59	the federal	prohibition	, from	prohib
iowa	108	108	ammunition is	prohibited	, that	prohib
iowa	329	329	ammunition is	prohibited	, and	prohib
iowa	386	386	not be	prohibited	from possessing	prohib
iowa	478	478	Requirement The	prohibition	on possessing	prohib
kansas	10	10	and Possession	Prohibitions	Kansas prohibits	prohib
kansas	12	12	Prohibitions Kansas	prohibits	possession of	prohib
kansas	59	59	1 Kansas	prohibits	possession of	prohib
kansas	175	175	terms explicitly	prohibits	the use	prohib
kentucky	10	10	and Possession	Prohibitions	Kentucky does	prohib
kentucky	14	14	does not	prohibit	purchase or	prohib
kentucky	33	33	does not	prohibit	purchase or	prohib
kentucky	198	198	direct or	prohibit	” any	prohib
kentucky	537	537	a person	prohibited	from purchasing	prohib
louisiana	10	10	and Possession	Prohibitions	Louisiana prohibits	prohib
louisiana	12	12	Prohibitions Louisiana	prohibits	persons convicted	prohib
louisiana	82	82	1 Louisiana	prohibits	the subjects	prohib
louisiana	170	170	person is	prohibited	from possessing	prohib
louisiana	322	322	… which	prohibits	the violent	prohib
louisiana	384	384	injunctions shall	prohibit	the violent	prohib
louisiana	542	542	an injunction	prohibiting	a spouse	prohib
louisiana	1547	1547	no longer	prohibited	from possessing	prohib
louisiana	1594	1594	no longer	prohibited	from possessing	prohib
louisiana	1633	1633	no longer	prohibited	from possessing	prohib
maine	10	10	and Possession	Prohibitions	Maine prohibits	prohib
maine	12	12	Prohibitions Maine	prohibits	persons from	prohib
maine	92	92	1 Maine	prohibits	persons from	prohib
maine	237	237	, explicitly	prohibits	the use	prohib
maryland	10	10	and Possession	Prohibitions	Maryland prohibits	prohib
maryland	12	12	Prohibitions Maryland	prohibits	persons convicted	prohib
maryland	51	51	1 Maryland	prohibits	persons convicted	prohib
maryland	68	68	2 Maryland	prohibits	persons convicted	prohib
maryland	144	144	does not	prohibit	the subject	prohib
maryland	168	168	Maryland may	prohibit	the subject	prohib
maryland	279	279	4 Maryland	prohibits	the subject	prohib
maryland	329	329	protective order	prohibits	the respondent	prohib
massachusetts	10	10	and Possession	Prohibitions	Massachusetts issues	prohib
michigan	10	10	and Possession	Prohibitions	Michigan does	prohib
michigan	14	14	does not	prohibit	purchase and	prohib
michigan	56	56	stalking may	prohibit	purchase and	prohib
minnesota	10	10	and Possession	Prohibitions	Minnesota prohibits	prohib
minnesota	12	12	Prohibitions Minnesota	prohibits	possession of	prohib
minnesota	60	60	Minnesota also	prohibits	possession of	prohib
minnesota	205	205	court may	prohibit	the person	prohib
minnesota	230	230	Minnesota also	prohibits	a person	prohib
minnesota	319	319	petitioner or	prohibits	the abusing	prohib
minnesota	366	366	abuse that	prohibits	possession of	prohib
minnesota	730	730	firearm possession	prohibition	shall determine	prohib
minnesota	1398	1398	of the	prohibiting	time period	prohib
minnesota	1409	1409	not otherwise	prohibited	from possessing	prohib
mississippi	10	10	and Possession	Prohibitions	Mississippi does	prohib
mississippi	14	14	does not	prohibit	purchase and	prohib
mississippi	35	35	does not	prohibit	purchase and	prohib
missouri	10	10	and Possession	Prohibitions	Missouri does	prohib
missouri	14	14	does not	prohibit	purchase and	prohib
missouri	35	35	does not	prohibit	purchase and	prohib
montana	10	10	and Possession	Prohibitions	Montana does	prohib
montana	15	15	not broadly	prohibit	the purchase	prohib
montana	37	37	court may	prohibit	an individual	prohib
montana	97	97	not broadly	prohibit	purchase and	prohib
montana	129	129	protection may	prohibit	” the	prohib
nebraska	10	10	and Possession	Prohibitions	Nebraska prohibits	prohib
nebraska	12	12	Prohibitions Nebraska	prohibits	possession of	prohib
nebraska	34	34	1 Nebraska	prohibits	possession of	prohib
nebraska	82	82	order may	prohibit	” the	prohib
nevada	10	10	and Possession	Prohibitions	Nevada prohibits	prohib
nevada	12	12	Prohibitions Nevada	prohibits	persons convicted	prohib
nevada	40	40	not explicitly	prohibit	the adverse	prohib
nevada	68	68	. Nevada	prohibits	the adverse	prohib
nevada	100	100	Nevada may	prohibit	the adverse	prohib
new-hampshire	12	12	and Possession	Prohibitions	New Hampshire	prohib
new-hampshire	17	17	does not	prohibit	persons convicted	prohib
new-hampshire	36	36	New Hampshire	prohibits	a defendant	prohib
new-hampshire	94	94	) may	prohibit	the defendant	prohib
new-jersey	12	12	and Possession	Prohibitions	New Jersey	prohib
new-jersey	15	15	New Jersey	prohibits	persons convicted	prohib
new-jersey	224	224	New Jersey	prohibits	persons subject	prohib
new-jersey	298	298	restraining order	prohibiting	the person	prohib
new-jersey	704	704	the order	prohibits	the defendant	prohib
new-jersey	852	852	the order	prohibits	the defendant	prohib
new-jersey	910	910	an order	prohibiting	the defendant	prohib
new-jersey	957	957	Requirement The	prohibition	on possession	prohib
new-jersey	1101	1101	court order	prohibiting	the possession	prohib
new-mexico	12	12	and Possession	Prohibitions	New Mexico	prohib
new-mexico	15	15	New Mexico	prohibits	purchase ,	prohib
new-york	12	12	and Possession	Prohibitions	Under New	prohib
north-carolina	12	12	and Possession	Prohibitions	North Carolina	prohib
north-carolina	17	17	does not	prohibit	purchase and	prohib
north-carolina	39	39	does not	prohibit	purchase and	prohib
north-carolina	65	65	order may	prohibit	the respondent	prohib
north-carolina	138	138	” is	prohibited	from possessing	prohib
north-carolina	352	352	any additional	prohibitions	or requirements	prohib
north-carolina	868	868	order that	prohibits	purchase and	prohib
north-carolina	983	983	not otherwise	prohibited	shall not	prohib
north-carolina	987	987	not be	prohibited	from possessing	prohib
north-carolina	1178	1178	order is	prohibited	under state	prohib
north-carolina	1315	1315	order is	prohibited	from regaining	prohib
north-dakota	12	12	and Possession	Prohibitions	Regardless of	prohib
north-dakota	26	26	North Dakota	prohibits	persons convicted	prohib
north-dakota	178	178	does not	prohibit	persons subject	prohib
ohio	10	10	and Possession	Prohibitions	Ohio does	prohib
ohio	14	14	does not	prohibit	purchase and	prohib
ohio	36	36	not explicitly	prohibit	purchase and	prohib
oklahoma	10	10	and Possession	Prohibitions	Oklahoma does	prohib
oklahoma	14	14	does not	prohibit	purchase or	prohib
oklahoma	35	35	does not	prohibit	purchase or	prohib
oklahoma	106	106	not specifically	prohibit	the defendant	prohib
oregon	10	10	and Possession	Prohibitions	Oregon prohibits	prohib
oregon	12	12	Prohibitions Oregon	prohibits	persons convicted	prohib
oregon	106	106	Oregon also	prohibits	persons convicted	prohib
oregon	120	120	3 Oregon	prohibits	possession of	prohib
pennsylvania	10	10	and Possession	Prohibitions	Pennsylvania prohibits	prohib
pennsylvania	12	12	Prohibitions Pennsylvania	prohibits	possession ,	prohib
pennsylvania	31	31	who are	prohibited	from possessing	prohib
pennsylvania	55	55	1 Pennsylvania	prohibits	possession ,	prohib
pennsylvania	139	139	Are otherwise	prohibited	from possessing	prohib
pennsylvania	1508	1508	of Pennsylvania’s	prohibition	on possession	prohib
pennsylvania	1567	1567	federal *	prohibition	on purchase	prohib
pennsylvania	1750	1750	of Pennsylvania’s	prohibition	on possession	prohib
pennsylvania	1821	1821	is not	prohibited	from possessing	prohib
pennsylvania	2183	2183	is not	prohibited	from possessing	prohib
pennsylvania	2461	2461	party is	prohibited	from possessing	prohib
pennsylvania	2811	2811	be otherwise	prohibited	by applicable	prohib
pennsylvania	3022	3022	is not	prohibited	from possessing	prohib
pennsylvania	3068	3068	defendant is	prohibited	from possessing	prohib
rhode-island	12	12	and Possession	Prohibitions	Rhode Island	prohib
rhode-island	15	15	Rhode Island	prohibits	persons who	prohib
rhode-island	67	67	Rhode Island	prohibits	persons subject	prohib
rhode-island	903	903	is not	prohibited	from possessing	prohib
rhode-island	1115	1115	the firearm	prohibition	if the	prohib
rhode-island	1174	1174	The firearm	prohibition	due solely	prohib
rhode-island	1350	1350	order that	prohibited	the person	prohib
rhode-island	1388	1388	not otherwise	prohibited	from possessing	prohib
south-carolina	12	12	and Possession	Prohibitions	South Carolina	prohib
south-carolina	15	15	South Carolina	prohibits	the shipping	prohib
south-carolina	93	93	person is	prohibited	from shipping	prohib
south-carolina	134	134	person is	prohibited	from shipping	prohib
south-carolina	283	283	South Carolina	prohibits	the shipping	prohib
south-carolina	408	408	person is	prohibited	from shipping	prohib
south-dakota	12	12	and Possession	Prohibitions	South Dakota	prohib
south-dakota	17	17	does not	prohibit	purchase or	prohib
south-dakota	39	39	does not	prohibit	purchase and	prohib
tennessee	10	10	and Possession	Prohibitions	Tennessee prohibits	prohib
tennessee	12	12	Prohibitions Tennessee	prohibits	possession of	prohib
tennessee	447	447	is not	prohibited	from possessing	prohib
tennessee	472	472	respondent is	prohibited	from possessing	prohib
tennessee	563	563	is not	prohibited	from possessing	prohib
texas	10	10	and Possession	Prohibitions	Texas prohibits	prohib
texas	12	12	Prohibitions Texas	prohibits	persons convicted	prohib
texas	157	157	4 Texas	prohibits	persons -	prohib
utah	10	10	and Possession	Prohibitions	Utah prohibits	prohib
utah	12	12	Prohibitions Utah	prohibits	persons convicted	prohib
utah	112	112	1 Utah	prohibits	persons subject	prohib
utah	179	179	, may	prohibit	the respondent	prohib
utah	236	236	may not	prohibit	the respondent	prohib
utah	791	791	an order	prohibiting	the respondent	prohib
vermont	10	10	and Possession	Prohibitions	Vermont prohibits	prohib
vermont	12	12	Prohibitions Vermont	prohibits	persons convicted	prohib
vermont	73	73	does not	prohibit	persons subject	prohib
vermont	243	243	court to	prohibit	the defendant	prohib
vermont	631	631	is not	prohibited	from owning	prohib
virginia	10	10	and Possession	Prohibitions	Virginia does	prohib
virginia	14	14	does not	prohibit	purchase or	prohib
virginia	33	33	. Virginia	prohibits	persons subject	prohib
virginia	171	171	Virginia also	prohibits	persons subject	prohib
virginia	258	258	abuse is	prohibited	from purchasing	prohib
virginia	319	319	not otherwise	prohibited	by law	prohib
washington	10	10	and Possession	Prohibitions	Washington prohibits	prohib
washington	12	12	Prohibitions Washington	prohibits	possession of	prohib
washington	90	90	1 Washington	prohibits	possession of	prohib
washington	235	235	, explicitly	prohibits	the use	prohib
washington	279	279	firearms and	prohibiting	the person	prohib
washington	391	391	] ;	Prohibit	the party	prohib
washington	407	407	weapons ;	Prohibit	the party	prohib
washington	434	434	firearm possession	prohibitory	under Washington	prohib
washington	474	474	] ;	Prohibit	the party	prohib
washington	490	490	weapons ;	Prohibit	the party	prohib
washington	599	599	] ;	Prohibit	the party	prohib
washington	615	615	weapons ;	Prohibit	the party	prohib
washington	770	770	] ;	Prohibit	the party	prohib
washington	786	786	weapons ;	Prohibit	the party	prohib
washington	1868	1868	this case	prohibiting	the restrained	prohib
washington	1916	1916	disqualifications that	prohibit	the restrained	prohib
washington	2016	2016	or otherwise	prohibited	from being	prohib
washington	2298	2298	a person	prohibited	due to	prohib
washington	2315	2315	and Possession	Prohibitions	” is	prohib
west-virginia	12	12	and Possession	Prohibitions	West Virginia	prohib
west-virginia	15	15	West Virginia	prohibits	persons convicted	prohib
west-virginia	110	110	West Virginia	prohibits	the subject	prohib
west-virginia	230	230	terms explicitly	prohibits	the use	prohib
west-virginia	262	262	West Virginia	prohibits	the respondent	prohib
west-virginia	282	282	not explicitly	prohibit	the subject	prohib
west-virginia	875	875	not otherwise	prohibited	by law	prohib
west-virginia	1037	1037	not otherwise	prohibited	by law	prohib
west-virginia	1255	1255	not otherwise	prohibited	from possessing	prohib
wisconsin	10	10	and Possession	Prohibitions	Wisconsin does	prohib
wisconsin	14	14	does not	prohibit	purchase or	prohib
wisconsin	33	33	. Wisconsin	prohibits	possession of	prohib
wisconsin	85	85	the firearm	prohibition	and penalties	prohib
wisconsin	982	982	is not	prohibited	from possessing	prohib
wisconsin	1023	1023	person is	prohibited	from possessing	prohib
wisconsin	1141	1141	person is	prohibited	from possessing	prohib
wisconsin	1184	1184	is not	prohibited	from possessing	prohib
wisconsin	1692	1692	person is	prohibited	from possessing	prohib
wisconsin	1759	1759	is not	prohibited	from possessing	prohib
wisconsin	1787	1787	is not	prohibited	from possessing	prohib
wisconsin	2015	2015	is not	prohibited	from possessing	prohib
wyoming	10	10	and Possession	Prohibitions	Wyoming does	prohib
wyoming	14	14	does not	prohibit	purchase or	prohib
wyoming	35	35	does not	prohibit	purchase or	prohib

kwic(KeyWords, pattern = "order", valuetype = "regex", window = 2) %>%
  kable() %>% kable_styling() %>% scroll_box(width = "100%", height = "400px")

docname	from	to	pre	keyword	post	pattern
alabama	50	50	valid protection	order	for domestic	order
alabama	59	59	Valid protection	order	” is	order
alabama	66	66	” an	order	issued after	order
alabama	204	204	Civil Protection	Orders	That Require	order
alabama	225	225	violence protection	orders	. Alabama	order
alabama	238	238	parte protection	order	, an	order
alabama	247	247	a protection	order	, a	order
alabama	252	252	final protection	order	, or	order
alabama	260	260	a protection	order	issued after	order
alabama	308	308	a Protection	Order	The following	order
alabama	317	317	a protection	order	: spouse	order
alabama	473	473	a protective	order	is a	order
alaska	20	20	a protective	order	issued after	order
alaska	100	100	violence protective	order	. Domestic	order
alaska	106	106	Civil Protective	Orders	That Require	order
alaska	127	127	violence protective	orders	; however	order
alaska	138	138	violence protective	order	, after	order
alaska	209	209	a Protective	Order	” household	order
alaska	219	219	a protective	order	. ”	order
alaska	366	366	violence protective	order	may ”	order
alaska	433	433	violence protective	order	. Penalties	order
alaska	442	442	a protective	order	may be	order
arizona	54	54	an emergency	order	of protection	order
arizona	67	67	of the	order	” [	order
arizona	102	102	domestic violence	order	of protection	order
arizona	120	120	of the	order	” [	order
arizona	156	156	Violence Civil	Orders	of Protection	order
arizona	171	171	domestic violence	order	of protection	order
arizona	187	187	” …	order	the defendant	order
arizona	223	223	an emergency	order	of protection	order
arizona	233	233	for an	Order	of Protection	order
arizona	245	245	domestic violence	order	of protection	order
arizona	296	296	or court	order	as a	order
arizona	399	399	Process defendant	ordered	to transfer	order
arizona	420	420	of the	order	to the	order
arizona	432	432	of the	order	. If	order
arizona	457	457	of the	order	. ”	order
arizona	467	467	of an	order	of protection	order
arkansas	40	40	to any	order	of protection	order
arkansas	47	47	Violence Civil	Orders	of Protection	order
arkansas	65	65	subject to	orders	of protection	order
arkansas	79	79	a temporary	order	of protection	order
arkansas	84	84	or final	order	of protection	order
arkansas	121	121	a Protective	Order	The following	order
arkansas	130	130	a protective	order	: Spouses	order
arkansas	188	188	of an	order	of protection	order
california	58	58	violence protective	order	whether issued	order
california	80	80	a protective	order	relating to	order
california	94	94	Civil Protective	Orders	That Require	order
california	101	101	Any protective	order	that includes	order
california	109	109	following restraining	orders	, whether	order
california	134	134	4 An	order	issued pursuant	order
california	148	148	Ex parte	order	enjoining contact	order
california	178	178	Ex parte	order	excluding party	order
california	195	195	Ex parte	order	enjoining specified	order
california	206	206	to effectuate	orders	under Section	order
california	222	222	a Protective	Order	The following	order
california	233	233	violence protective	orders	: 6	order
california	398	398	violence protective	order	, the	order
california	464	464	prior protective	order	or any	order
california	472	472	prior protective	order	. 7	order
california	482	482	a protective	order	, listed	order
california	490	490	court shall	order	the respondent	order
california	517	517	The protective	order	shall also	order
california	534	534	of the	order9	and the	order
california	537	537	and the	order	” shall	order
california	563	563	the protective	order	is in	order
california	612	612	of the	order	. ”	order
california	622	622	a protective	order	that indicates	order
california	669	669	with the	order	. 12	order
california	745	745	the protective	order	and a	order
california	761	761	the protective	order	within 48	order
california	769	769	of the	order	. 15	order
california	783	783	a protective	order	has a	order
california	791	791	of such	order	, the	order
california	816	816	the protective	order	has a	order
california	824	824	of such	order	” at	order
california	835	835	violence protective	order	is issued	order
california	856	856	while the	order	remains in	order
california	875	875	a protective	order	has a	order
california	883	883	of such	order	in issuing	order
california	888	888	: An	order	to show	order
california	897	897	or An	order	for money	order
california	919	919	with the	order	, California	order
california	1137	1137	limit the	order	to exclude	order
california	1160	1160	the protective	order	is in	order
california	1182	1182	a protective	order	based on	order
california	1250	1250	a protective	order	, though	order
california	1298	1298	a protective	order	. 26	order
california	1314	1314	of an	order	, the	order
california	1361	1361	violence protective	order	is issued	order
california	1584	1584	a protective	order	” is	order
california	1643	1643	a protective	order	shall constitute	order
california	1690	1690	for an	order	above ,	order
california	1701	1701	harassment restraining	order	, workplace	order
california	1706	1706	violence restraining	order	, or	order
california	1715	1715	abuse restraining	order	.	order
colorado	46	46	civil protection	order	that qualifies	order
colorado	51	51	as an	order	pursuant to	order
colorado	82	82	civil protection	order	( which	order
colorado	97	97	Civil Protection	Orders	that Require	order
colorado	109	109	civil protection	order	that qualifies	order
colorado	114	114	as an	order	pursuant to	order
colorado	136	136	) to	order	the subject	order
colorado	143	143	civil protection	order	to ”	order
colorado	179	179	civil protection	order	may ”	order
colorado	182	182	may ”	order	any other	order
colorado	199	199	civil protection	order	may grant	order
colorado	223	223	Violence Protection	Order	The following	order
colorado	233	233	aforementioned protection	order	: current	order
colorado	259	259	of the	order	; current	order
colorado	284	284	civil protection	order	, the	order
colorado	319	319	of the	order	; current	order
colorado	334	334	civil protection	order	that qualifies	order
colorado	339	339	as an	order	pursuant to	order
colorado	361	361	) shall	order	the subject	order
colorado	368	368	civil protection	order	to ”	order
colorado	385	385	of the	order	; and	order
colorado	395	395	of the	order	, any	order
colorado	426	426	civil protection	order	is required	order
colorado	443	443	with the	order	in open	order
colorado	456	456	with the	order	outside of	order
colorado	854	854	the protection	order	and the	order
colorado	961	961	a protection	order	is a	order
colorado	979	979	a protection	order	or analogous	order
connecticut	225	225	or protective	order	issued by	order
connecticut	413	413	or protective	order	( issued	order
connecticut	423	423	a foreign	order	of protection	order
connecticut	542	542	or Protective	Orders	that Require	order
connecticut	570	570	or Protective	Order	” Any	order
connecticut	623	623	or protective	order	. 11	order
connecticut	712	712	or protective	order	for which	order
connecticut	738	738	of the	order	: Sell	order
connecticut	757	757	of the	order	possesses to	order
connecticut	782	782	of the	order	possesses to	order
connecticut	856	856	a restraining	order	or a	order
connecticut	860	860	a protective	order	, or	order
connecticut	919	919	the restraining	order	or protective	order
connecticut	922	922	or protective	order	, the	order
connecticut	979	979	or protective	order	who surrendered	order
connecticut	1026	1026	the restraining	order	or protective	order
connecticut	1029	1029	or protective	order	; or	order
connecticut	1037	1037	a court	order	that rescinds	order
connecticut	1042	1042	the restraining	order	or protective	order
connecticut	1045	1045	or protective	order	. 16	order
connecticut	1062	1062	the restraining	order	or protective	order
connecticut	1065	1065	or protective	order	or the	order
connecticut	1072	1072	a court	order	rescinding the	order
connecticut	1076	1076	the restraining	order	or protective	order
connecticut	1079	1079	or protective	order	to the	order
connecticut	1145	1145	the restraining	order	or protective	order
connecticut	1148	1148	or protective	order	has expired	order
connecticut	1158	1158	issued an	order	that rescinds	order
connecticut	1163	1163	the restraining	order	or protective	order
connecticut	1166	1166	or protective	order	; That	order
connecticut	1228	1228	a restraining	order	or protective	order
connecticut	1231	1231	or protective	order	who fails	order
delaware	54	54	from abuse	order	( other	order
delaware	61	61	ex parte	order	) .	order
delaware	75	75	from abuse	order	may ”	order
delaware	78	78	may ”	order	the respondent	order
delaware	95	95	of the	order	. The	order
delaware	123	123	the protective	order	is in	order
delaware	136	136	from Abuse	Orders	that Require	order
delaware	157	157	from abuse	order	; 4	order
delaware	173	173	from abuse	order	or a	order
delaware	179	179	from abuse	order	after notice	order
delaware	185	185	hearing may	order	firearms and	order
delaware	205	205	from Abuse	Order	The following	order
delaware	216	216	from abuse	order	: Family	order
delaware	402	402	from abuse	order	may :	order
delaware	406	406	: ”	Order	the respondent	order
delaware	441	441	of the	order	… ”	order
delaware	447	447	Issue an	order	directing any	order
delaware	580	580	from abuse	order	. The	order
delaware	619	619	from abuse	order	requiring the	order
delaware	666	666	from abuse	order	. If	order
delaware	694	694	from abuse	order	at any	order
delaware	781	781	from abuse	order	or ,	order
delaware	805	805	from abuse	order	, within	order
delaware	853	853	of the	order	and does	order
delaware	892	892	of the	order	, that	order
delaware	940	940	of the	order	, that	order
delaware	987	987	from abuse	order	shall the	order
delaware	1023	1023	from abuse	order	may be	order
delaware	1273	1273	from abuse	order	and that	order
delaware	1287	1287	from abuse	order	shall arrest	order
delaware	1306	1306	from abuse	order	, in	order
delaware	1354	1354	from abuse	order	exists .	order
delaware	1363	1363	from abuse	order	is not	order
delaware	1389	1389	from abuse	order	exists .	order
delaware	1405	1405	from abuse	order	cannot be	order
delaware	1420	1420	with the	order	shall inform	order
delaware	1427	1427	of the	order	, make	order
delaware	1439	1439	with the	order	, and	order
delaware	1452	1452	with the	order	before enforcing	order
delaware	1456	1456	enforcing the	order	. 19	order
delaware	1466	1466	from abuse	order	is punishable	order
florida	124	124	courts to	order	the removal	order
florida	161	161	a foreign	order	of protection	order
florida	177	177	ammunition if	ordered	to do	order
florida	261	261	violence may	order	” such	order
georgia	40	40	family violence	orders	from purchasing	order
georgia	52	52	Family Violence	Orders	that Require	order
georgia	71	71	family violence	orders	; however	order
georgia	83	83	to ”	order	such temporary	order
georgia	115	115	any protective	order	… to	order
georgia	137	137	a Protective	Order	The following	order
georgia	148	148	violence protective	order	: Past	order
georgia	196	196	violence protective	order	may be	order
hawaii	131	131	a court	order	, including	order
hawaii	137	137	ex parte	order	, from	order
hawaii	153	153	as the	order	is in	order
hawaii	174	174	Civil Restraining	Orders	and Protective	order
hawaii	177	177	and Protective	Orders	that Require	order
hawaii	201	201	a court	order	, including	order
hawaii	207	207	ex parte	order	, from	order
hawaii	230	230	Violence Restraining	Order	or Protective	order
hawaii	233	233	or Protective	Order	Any of	order
hawaii	245	245	temporary restraining	order	or a	order
hawaii	250	250	final protective	order	: Spouses	order
hawaii	334	334	disqualifying restraining	orders	or protective	order
hawaii	337	337	or protective	orders	shall ,	order
hawaii	381	381	such an	order	, the	order
hawaii	388	388	serving the	order	may take	order
hawaii	423	423	of the	order	. 7	order
hawaii	445	445	of the	order	or known	order
hawaii	574	574	or protective	order	who has	order
hawaii	585	585	of the	order	shall be	order
idaho	41	41	violence protection	orders	from purchasing	order
idaho	54	54	Civil Protection	Orders	that Require	order
idaho	73	73	violence protective	orders	; however	order
idaho	112	112	violence protection	order	, issued	order
idaho	121	121	, may	order	” [	order
idaho	145	145	, including	orders	or directives	order
idaho	173	173	violence protection	order	, may	order
idaho	176	176	, may	order	” other	order
idaho	196	196	, including	orders	or directives	order
idaho	221	221	a Protection	Order	The following	order
idaho	230	230	a protection	order	: Spouses	order
idaho	288	288	the protection	order	had notice	order
idaho	293	293	of the	order	, a	order
idaho	300	300	a protection	order	shall be	order
illinois	93	93	of an	order	of protection	order
illinois	156	156	a civil	order	of protection	order
illinois	161	161	, interim	order	of protection	order
illinois	167	167	or plenary	order	of protection	order
illinois	181	181	whom an	order	of protection	order
illinois	195	195	of the	order	if the	order
illinois	198	198	if the	order	: was	order
illinois	335	335	civil emergency	order	of protection	order
illinois	450	450	an existing	order	of protection	order
illinois	458	458	Violence Civil	Orders	of Protection	order
illinois	470	470	a civil	order	of protection	order
illinois	475	475	, interim	order	of protection	order
illinois	481	481	or plenary	order	of protection	order
illinois	496	496	also :	Order	the respondent	order
illinois	537	537	for an	Order	of Protection	order
illinois	547	547	for an	order	of protection	order
illinois	862	862	by court	order	. ”	order
illinois	874	874	a civil	order	of protection	order
illinois	879	879	, interim	order	of protection	order
illinois	885	885	or plenary	order	of protection	order
illinois	899	899	shall also	order	the respondent	order
illinois	947	947	of the	order	. The	order
illinois	983	983	of the	order	. 13	order
illinois	997	997	court shall	order	the respondent	order
illinois	1037	1037	of the	order	. 14	order
illinois	1050	1050	of the	order	, the	order
illinois	1135	1135	court may	order	( a	order
illinois	1216	1216	violates an	order	of protection	order
illinois	1255	1255	of the	order	. 17	order
illinois	1261	1261	of an	order	of protection	order
illinois	1273	1273	of an	order	of protection	order
illinois	1296	1296	of an	order	of protection	order
illinois	1324	1324	a protection	order	. 19	order
illinois	1330	1330	of an	order	of protection	order
indiana	29	29	domestic violence	order	for protection	order
indiana	70	70	Violence Civil	Orders	for Protection	order
indiana	90	90	domestic violence	orders	for protection	order
indiana	102	102	domestic violence	order	for protection	order
indiana	111	111	hearing may	order	a respondent	order
indiana	130	130	ex parte	order	for protection	order
indiana	166	166	for an	Order	for Protection	order
indiana	300	300	domestic violence	order	for protection	order
indiana	334	334	of the	order	for protection	order
indiana	341	341	date is	ordered	by the	order
indiana	356	356	judge may	order	” a	order
indiana	415	415	of the	order	of protection	order
indiana	422	422	date is	ordered	by the	order
indiana	438	438	domestic violence	order	for protection	order
iowa	52	52	violence protective	order	, that	order
iowa	80	80	Civil Protective	Orders	that Require	order
iowa	91	91	violence protective	order	, for	order
iowa	116	116	of the	order	to be	order
iowa	127	127	ammunition shall	order	that such	order
iowa	150	150	a Protective	Order	The following	order
iowa	161	161	violence protective	order	: Family	order
iowa	311	311	violence protective	order	, for	order
iowa	338	338	of the	order	to be	order
iowa	349	349	ammunition shall	order	that such	order
iowa	419	419	court shall	order	that the	order
iowa	462	462	until such	order	is no	order
iowa	500	500	violence protective	order	under 18	order
iowa	572	572	the protective	order	is no	order
iowa	626	626	violence protective	order	under 18	order
kansas	72	72	a court	order	that :	order
kansas	212	212	from Abuse	Orders	that Require	order
kansas	232	232	from abuse	orders	; however	order
kansas	248	248	from abuse	order	to ”	order
kansas	251	251	to ”	order	[ ]	order
kansas	290	290	from Abuse	Order	The following	order
kansas	301	301	from abuse	order	: Persons	order
kansas	348	348	from abuse	order	is a	order
kansas	389	389	from abuse	order	may find	order
kentucky	46	46	violence protective	orders	. Kentucky	order
kentucky	65	65	domestic violence	order	or emergency	order
kentucky	69	69	emergency protective	order	. 1	order
kentucky	81	81	of the	order	or ”	order
kentucky	90	90	issued the	order	terminates the	order
kentucky	103	103	domestic violence	order	or emergency	order
kentucky	107	107	emergency protective	order	[ .	order
kentucky	117	117	Violence Protective	Orders	that Require	order
kentucky	135	135	domestic violence	orders	or emergency	order
kentucky	139	139	emergency protective	orders	. Kentucky	order
kentucky	151	151	domestic violence	order	or emergency	order
kentucky	155	155	emergency protective	order	to surrender	order
kentucky	178	178	served the	order	. 3	order
kentucky	189	189	emergency protective	order	or a	order
kentucky	194	194	domestic violence	order	may direct	order
kentucky	231	231	include an	order	not to	order
kentucky	238	238	and an	order	to relinquish	order
kentucky	251	251	a Protective	Order	victim of	order
kentucky	264	264	emergency protective	order	or domestic	order
kentucky	268	268	domestic violence	order	or any	order
kentucky	438	438	and protective	order	history prior	order
kentucky	452	452	a protective	order	and use	order
kentucky	485	485	of an	order	of protection	order
kentucky	500	500	of the	order	, shall	order
kentucky	577	577	domestic violence	order	if the	order
louisiana	91	91	and protective	orders	listed below	order
louisiana	106	106	or protective	order	if the	order
louisiana	112	112	or protective	order	includes (	order
louisiana	127	127	or protective	order	represents a	order
louisiana	156	156	or protective	order	informs the	order
louisiana	165	165	or protective	order	that the	order
louisiana	198	198	violence protective	order	; Injunction	order
louisiana	212	212	violence protective	order	. Civil	order
louisiana	218	218	Violence Protective	Orders	that Require	order
louisiana	232	232	violence protective	order	, an	order
louisiana	247	247	violence protective	order	” the	order
louisiana	252	252	judge shall	order	the transfer	order
louisiana	276	276	injunction or	order	[ .	order
louisiana	291	291	or Protective	Order	Post-separation family	order
louisiana	296	296	violence protective	order	” All	order
louisiana	309	309	child visitation	orders	and judgments	order
louisiana	442	442	for court	ordered	visitation or	order
louisiana	567	567	violence protective	order	An adult	order
louisiana	582	582	violence7 protective	order	. Any	order
louisiana	601	601	violence protective	order	on behalf	order
louisiana	819	819	violence protective	order	, an	order
louisiana	834	834	violence protective	order	” the	order
louisiana	839	839	judge shall	order	the transfer	order
louisiana	863	863	injunction or	order	[ .	order
louisiana	874	874	of such	order	, the	order
louisiana	957	957	of the	order	to transfer	order
louisiana	991	991	court shall	order	the defendant	order
louisiana	1003	1003	of the	order	, to	order
louisiana	1022	1022	of the	order	and firearm	order
louisiana	1120	1120	which the	order	was issued	order
louisiana	1568	1568	seeking an	order	for the	order
louisiana	1608	1608	court shall	order	the transferred	order
louisiana	1621	1621	30 Such	order	shall include	order
louisiana	1670	1670	receiving an	order	from the	order
louisiana	1691	1691	violence protective	order	, an	order
louisiana	1706	1706	violence protective	order	, ”	order
louisiana	1728	1728	the protective	order	, the	order
louisiana	1769	1769	violence protective	order	, an	order
louisiana	1784	1784	violence protective	order	, ”	order
louisiana	1806	1806	the protective	order	, regardless	order
louisiana	1866	1866	violence protective	order	, an	order
louisiana	1881	1881	violence protective	order	” where	order
louisiana	1903	1903	the protective	order	is in	order
louisiana	1950	1950	violence protective	order	, an	order
louisiana	1965	1965	violence protective	order	” where	order
louisiana	1987	1987	the protective	order	is in	order
louisiana	2001	2001	a protective	order	or of	order
louisiana	2016	2016	the protective	order	is in	order
maine	112	112	to an	order	of a	order
maine	271	271	from Abuse	Orders	that Require	order
maine	282	282	from abuse	order	may ”	order
maine	310	310	of the	order	if the	order
maine	377	377	The temporary	order	of protection	order
maine	399	399	has violated	orders	of protection	order
maine	470	470	the court	orders	the defendant	order
maine	527	527	from abuse	order	may direct	order
maine	554	554	of the	order	[ . ] if	order
north-carolina	343	343	a protective	order	may ”	order
north-carolina	379	379	Violence Protective	Order	If the	order
north-carolina	399	399	a protective	order	. 5	order
north-carolina	596	596	obtain an	order	of protection	order
north-carolina	702	702	the protective	order	, the	order
north-carolina	866	866	a protective	order	that prohibits	order
north-carolina	899	899	sheriff as	ordered	by the	order
north-carolina	977	977	a protective	order	and who	order
north-carolina	1029	1029	a protective	order	. ”	order
north-carolina	1047	1047	a protective	order	or final	order
north-carolina	1064	1064	a protective	order	and not	order
north-carolina	1077	1077	a protective	order	or final	order
north-carolina	1094	1094	a protective	order	, the	order
north-carolina	1100	1100	of the	order	may make	order
north-carolina	1176	1176	the protective	order	is prohibited	order
north-carolina	1201	1201	the protective	order	has been	order
north-carolina	1212	1212	the protective	order	is subject	order
north-carolina	1219	1219	other protective	orders	; Whether	order
north-carolina	1227	1227	the protective	order	is disqualified	order
north-carolina	1249	1249	the protective	order	has any	order
north-carolina	1273	1273	current protective	order	. If	order
north-carolina	1281	1281	a protective	order	does not	order
north-carolina	1313	1313	the protective	order	is prohibited	order
north-carolina	1335	1335	the protective	order	fails to	order
north-carolina	1358	1358	of the	order	granting return	order
north-carolina	1390	1390	of the	order	and apply	order
north-carolina	1398	1398	for an	order	of disposition	order
north-carolina	1410	1410	, may	order	the disposition	order
north-carolina	1435	1435	: By	ordering	the weapon	order
north-carolina	1482	1482	; By	ordering	the weapon	order
north-carolina	1528	1528	; By	ordering	the weapon	order
north-carolina	1553	1553	; By	ordering	the weapon	order
north-carolina	1615	1615	a protective	order	shall be	order
north-dakota	185	185	violence protection	orders	from purchasing	order
north-dakota	198	198	Civil Protection	Orders	That Require	order
north-dakota	224	224	violence protection	orders	; however	order
north-dakota	239	239	violence protection	order	may require	order
north-dakota	321	321	Violence Protection	Order	Any family	order
north-dakota	337	337	violence protection	order	. 3	order
north-dakota	420	420	the court	orders	the respondent	order
north-dakota	547	547	violence protection	order	has been	order
north-dakota	562	562	of the	order	, the	order
north-dakota	569	569	of an	order	is a	order
north-dakota	604	604	of an	order	is a	order
ohio	51	51	violence protection	orders	, though	order
ohio	58	58	violence protection	order	forms *	order
ohio	84	84	while the	order	remains in	order
ohio	97	97	Violence Protection	Orders	that Require	order
ohio	118	118	violence protection	orders	; however	order
ohio	129	129	temporary protection	order	that includes	order
ohio	152	152	final protection	order	may also	order
ohio	177	177	to ,	ordering	the respondent	order
ohio	229	229	violence protection	order	forms *	order
ohio	273	273	with this	order	as follows	order
ohio	314	314	a Protection	Order	person who	order
ohio	360	360	violence protection	order	. 5	order
ohio	718	718	a protection	order	is subject	order
ohio	731	731	a protection	order	which is	order
ohio	779	779	a protection	order	or consent	order
ohio	824	824	the protection	order	or consent	order
oklahoma	50	50	violence protective	orders	. An	order
oklahoma	58	58	final protective	order	shall state	order
oklahoma	83	83	while an	order	is in	order
oklahoma	102	102	if the	order	does not	order
oklahoma	123	123	Abuse Protective	Orders	that Require	order
oklahoma	143	143	violence protective	orders	; however	order
oklahoma	156	156	ex parte	order	that it	order
oklahoma	184	184	final protective	order	may ”	order
oklahoma	195	195	the protective	order	that the	order
oklahoma	236	236	Abuse Protective	Order	victim of	order
oklahoma	248	248	abuse protective	order	. 4	order
oklahoma	508	508	final protective	order	who is	order
oklahoma	515	515	of such	order	, upon	order
oklahoma	576	576	final protective	order	shall ,	order
oklahoma	633	633	final protective	order	who violated	order
oklahoma	638	638	the protective	order	and causes	order
oklahoma	656	656	in the	order	shall ,	order
oklahoma	706	706	final protective	order	which causes	order
oklahoma	724	724	in the	order	shall be	order
oregon	133	133	a court	order	that (	order
oregon	322	322	Act Restraining	Orders	that Require	order
oregon	348	348	) restraining	orders	; however	order
oregon	358	358	ex parte	order	shall ,	order
oregon	367	367	petitioner ,	order	” [	order
oregon	416	416	a Restraining	Order	Any person	order
oregon	436	436	a restraining	order	. 7	order
oregon	611	611	a restraining	order	; true	order
oregon	617	617	of the	order	and proof	order
oregon	651	651	of the	order	. 10	order
pennsylvania	83	83	from abuse	order	issued pursuant	order
pennsylvania	102	102	from abuse	order	issued pursuant	order
pennsylvania	118	118	if the	order	requires relinquishment	order
pennsylvania	134	134	of the	order	; or	order
pennsylvania	164	164	from Abuse	Orders	that Require	order
pennsylvania	177	177	from abuse	order	may ”	order
pennsylvania	203	203	the temporary	order	if the	order
pennsylvania	267	267	the temporary	order	of protection	order
pennsylvania	297	297	from abuse	order	. Whether	order
pennsylvania	387	387	from abuse	order	must order	order
pennsylvania	389	389	order must	order	the defendant	order
pennsylvania	448	448	agreement may	order	the defendant	order
pennsylvania	512	512	from Abuse	Order	An adult	order
pennsylvania	547	547	from abuse	order	. 6	order
pennsylvania	734	734	from abuse	order	shall relinquish	order
pennsylvania	758	758	the final	order	to :	order
pennsylvania	787	787	the court	orders	the subject	order
pennsylvania	798	798	from abuse	order	to relinquish	order
pennsylvania	822	822	the temporary	order	to :	order
pennsylvania	881	881	the court’s	order	for the	order
pennsylvania	887	887	of the	order	or until	order
pennsylvania	894	894	by court	order	” and	order
pennsylvania	937	937	from abuse	order	may relinquish	order
pennsylvania	1005	1005	from abuse	order	was issued	order
pennsylvania	1117	1117	from abuse	order	… which	order
pennsylvania	1120	1120	… which	order	provides for	order
pennsylvania	1187	1187	in the	order	. 14	order
pennsylvania	1198	1198	from abuse	order	, a	order
pennsylvania	1220	1220	a temporary	order	may request	order
pennsylvania	1283	1283	court shall	order	the sheriff	order
pennsylvania	1315	1315	from abuse	order	who wishes	order
pennsylvania	1351	1351	where the	order	was entered	order
pennsylvania	1361	1361	in the	order	. 17	order
pennsylvania	1403	1403	from abuse	order	was issued	order
pennsylvania	1534	1534	from abuse	order	and the	order
pennsylvania	1559	1559	from abuse	order	. plain-language	order
pennsylvania	1583	1583	violence protective	order	. 18	order
pennsylvania	1645	1645	from abuse	order	was issued	order
pennsylvania	1776	1776	from abuse	order	and the	order
pennsylvania	1801	1801	from abuse	order	. plain-language	order
pennsylvania	1854	1854	from abuse	order	. An	order
pennsylvania	1871	1871	from abuse	order	issued on	order
pennsylvania	2345	2345	in the	order	. ”	order
pennsylvania	2395	2395	where the	order	was entered	order
pennsylvania	2448	2448	from abuse	order	has been	order
pennsylvania	2534	2534	from abuse	order	consistent with	order
pennsylvania	2577	2577	from abuse	order	which requires	order
pennsylvania	2610	2610	a temporary	order	may have	order
pennsylvania	2646	2646	from abuse	order	requiring the	order
pennsylvania	2676	2676	of the	order	or dismissal	order
pennsylvania	2687	2687	from abuse	order	. 26	order
pennsylvania	2759	2759	from abuse	order	has been	order
pennsylvania	2882	2882	from abuse	order	. Such	order
pennsylvania	2980	2980	from abuse	order	which required	order
pennsylvania	3159	3159	the time	ordered	by the	order
pennsylvania	3236	3236	the protection	order	, or	order
pennsylvania	3303	3303	from abuse	orders	, except	order
pennsylvania	3397	3397	from abuse	order	that requires	order
pennsylvania	3462	3462	from abuse	order	may be	order
rhode-island	41	41	a protective	order	, and	order
rhode-island	44	44	, and	disorderly	conduct *	order
rhode-island	74	74	abuse protective	orders	and persons	order
rhode-island	82	82	assault protective	orders	issued after	order
rhode-island	113	113	Assault Protective	Orders	that Require	order
rhode-island	128	128	requesting any	order	which will	order
rhode-island	167	167	. further	order	[ ing ] ,	order	the respondent	order
wisconsin	1793	1793	firearm .	Order	the respondent	order
wisconsin	1835	1835	respondent is	ordered	to surrender	order
wisconsin	1882	1882	to the	order	, the	order
wisconsin	1893	1893	violating the	order	and the	order
wisconsin	1930	1930	the sheriff	ordering	that the	order
wisconsin	2029	2029	by the	order	of any	order
wisconsin	2067	2067	violates an	order	to surrender	order
wyoming	47	47	domestic violence	orders	of protection	order
wyoming	54	54	Domestic Violence	Orders	of Protection	order
wyoming	77	77	domestic violence	orders	of protection	order
wyoming	87	87	issuing an	order	of protection	order
wyoming	121	121	for an	Order	of Protection	order
wyoming	134	134	ex parte	order	of protection	order
wyoming	138	138	protection or	order	of protection	order
wyoming	338	338	ex parte	order	of protection	order
wyoming	342	342	protection or	order	of protection	order

Based on the patterns above, I will remove words in all caps, as those are headlines across all documents and provide no insight to the details of specific document provisions.

# Creat tri-grams to see if there are any noteworthy word combinations I'll want to compound
d <- tibble(disarmData)
trigram <- d %>%
 unnest_tokens(ngram, Text, token = "ngrams", n = 3)
trigrams_separated <- trigram %>%
  separate(ngram, c("word1", "word2", "word3"), sep = " ")
# Look at top 200 trigrams 
trigrams_counts <- trigrams_separated %>%
  count(word1, word2, word3, sort = TRUE) 
kable(head(trigrams_counts, 200)) %>% kable_styling() %>% 
  scroll_box(width = "100%", height = "400px")

word1	word2	word3	n
order	of	protection	129
may	petition	for	92
family	or	household	89
law	enforcement	agency	89
of	the	order	89
domestic	violence	firearm	88
domestic	violence	protective	84
protection	from	abuse	81
violence	protective	order	80
from	abuse	order	73
petition	for	a	71
purchase	and	possession	66
a	court	issuing	63
a	domestic	violence	63
or	household	member	63
a	protective	order	62
persons	subject	to	61
the	court	shall	61
the	subject	of	57
firearm	purchase	and	52
is	defined	as	52
and	possession	prohibitions	51
the	third	party	51
violence	firearm	purchase	51
of	domestic	violence	49
firearms	or	ammunition	48
individuals	who	may	47
an	ex	parte	46
an	order	of	46
a	law	enforcement	45
penalties	for	violation	45
who	may	petition	45
firearm	or	ammunition	44
subject	to	a	44
persons	convicted	of	43
firearms	and	ammunition	42
of	the	following	42
a	person	who	41
court	issuing	a	41
domestic	violence	protection	41
possessing	a	firearm	41
the	respondent	to	41
firearms	other	weapons	40
licensed	firearms	dealer	40
that	require	removal	40
a	firearm	or	38
possession	of	firearms	38
the	protective	order	38
for	a	domestic	37
a	child	in	36
final	domestic	violence	36
violence	firearm	prohibitions	36
weapons	or	ammunition	36
if	the	court	35
orders	that	require	35
child	in	common	34
not	more	than	34
order	firearm	removal	34
from	possessing	a	33
law	enforcement	officer	33
of	firearms	or	33
other	weapons	or	33
removal	of	firearms	33
a	fine	of	32
persons	who	have	32
prohibited	from	possessing	32
subject	of	the	32
that	the	respondent	32
a	family	or	31
federally	licensed	firearms	31
local	law	enforcement	31
the	law	enforcement	31
a	domestic	abuse	30
a	violation	of	30
duration	of	the	30
for	the	duration	30
order	for	protection	30
the	duration	of	30
all	firearms	and	29
any	firearm	or	29
does	not	prohibit	29
domestic	violence	order	29
if	the	respondent	29
protective	order	or	29
return	of	firearms	29
the	following	persons	29
civil	domestic	violence	28
firearm	or	other	28
firearms	ammunition	or	28
the	court	may	28
a	protection	from	27
at	the	time	27
concealed	pistol	license	27
for	violation	a	27
of	a	firearm	27
person	who	is	27
petition	for	an	27
the	respondent	is	27
violation	of	a	27
a	dating	relationship	26
does	not	require	26
law	domestic	violence	26
possession	of	the	26
the	abusing	party	26
violence	protection	order	26
by	the	respondent	25
firearm	removal	civil	25
or	protective	order	25
persons	who	are	25
ammunition	by	persons	24
court	issuing	an	24
member	of	the	24
or	ammunition	by	24
or	other	dangerous	24
orders	of	protection	24
pursuant	to	a	24
subject	to	domestic	24
that	the	person	24
who	has	been	24
current	or	former	23
domestic	abuse	protection	23
ex	parte	or	23
for	protection	against	23
is	a	class	23
on	behalf	of	23
possession	or	control	23
the	time	of	23
violation	of	an	23
a	third	party	22
any	of	the	22
from	possessing	firearms	22
is	in	effect	22
of	the	firearm	22
order	the	respondent	22
protective	order	the	22
related	by	blood	22
shall	order	the	22
the	respondent	has	22
to	domestic	violence	22
to	the	defendant	22
to	the	sheriff	22
by	the	court	21
convicted	of	a	21
following	persons	may	21
of	firearms	to	21
or	final	domestic	21
other	dangerous	weapons	21
parte	or	final	21
possess	a	firearm	21
possession	of	a	21
protective	order	is	21
temporary	or	final	21
weapon	or	ammunition	21
and	all	firearms	20
any	and	all	20
convicted	of	misdemeanor	20
final	protection	from	20
firearm	removal	domestic	20
for	an	order	20
of	any	firearm	20
of	protection	or	20
persons	may	petition	20
require	removal	of	20
the	firearm	s	20
to	a	domestic	20
to	the	respondent	20
a	federally	licensed	19
a	protection	order	19
a	temporary	or	19
abuse	is	defined	19
abuse	protection	order	19
any	person	who	19
cause	to	believe	19
from	persons	subject	19
have	a	child	19
of	an	order	19
or	other	weapon	19
or	possessing	firearms	19
protective	order	may	19
the	order	of	19
against	domestic	violence	18
crimes	of	domestic	18
domestic	violence	civil	18
domestic	violence	restraining	18
from	purchasing	or	18
guilty	of	a	18
in	the	same	18
issuing	an	ex	18
member	is	defined	18
misdemeanor	crimes	of	18
not	prohibit	purchase	18
not	require	removal	18
of	misdemeanor	crimes	18
of	not	more	18
or	other	weapons	18
or	possession	of	18
order	is	in	18
persons	related	by	18
protection	against	domestic	18
removal	civil	domestic	18

Based on the trigrams and reviewing the kwic() results, I will create compounded tokens for these phrases:

• “does not prohibit”
• “may prohibit”
• “shall prohibit”
• “not broadly prohibit”
• “not explicitly prohibit”
• “not specifically prohibit”
• “not require the removal”
• “require firearm removal”
• “does not require removal”
• “that require removal”
• “removal process”
• “exemption from removal requirement”
• “requires removal”
• “protective order”
• “restraining order”
• “temporary order”
• “temporary order”
• “order may”
• “order shall”
• “shall order”
• “may order”
• “law enforcement”
• “may petition for”
• “dating relationship”
• “child in common”
• “current or former”
• “domestic violence”
• “ex parte”
• “family or household member”

# Remove punctuation, separators, numbers, and symbols
# Change all remaining tokens to lowercase 
TextTokens <- tokens(Meta_corpus, remove_punct = TRUE,  remove_separators = TRUE, remove_numbers = TRUE, remove_symbols = TRUE) %>% tokens_tolower()
TextTokens

Tokens consisting of 50 documents and 8 docvars.
alabama :
 [1] "alabama"      "law"          "alabama"      "domestic"    
 [5] "violence"     "firearm"      "purchase"     "and"         
 [9] "possession"   "prohibitions" "alabama"      "prohibits"   
[ ... and 412 more ]

alaska :
 [1] "alaska"       "law"          "alaska"       "domestic"    
 [5] "violence"     "firearm"      "purchase"     "and"         
 [9] "possession"   "prohibitions" "alaska"       "may"         
[ ... and 399 more ]

arizona :
 [1] "arizona"      "law"          "arizona"      "domestic"    
 [5] "violence"     "firearm"      "purchase"     "and"         
 [9] "possession"   "prohibitions" "arizona"      "prohibits"   
[ ... and 413 more ]

arkansas :
 [1] "arkansas"     "law"          "arkansas"     "domestic"    
 [5] "violence"     "firearm"      "purchase"     "and"         
 [9] "possession"   "prohibitions" "arkansas"     "does"        
[ ... and 179 more ]

california :
 [1] "california"   "law"          "california"   "domestic"    
 [5] "violence"     "firearm"      "purchase"     "and"         
 [9] "possession"   "prohibitions" "california"   "prohibits"   
[ ... and 1,503 more ]

colorado :
 [1] "colorado"     "law"          "colorado"     "domestic"    
 [5] "violence"     "firearm"      "purchase"     "and"         
 [9] "possession"   "prohibitions" "colorado"     "prohibits"   
[ ... and 836 more ]

[ reached max_ndoc ... 44 more documents ]

# Remove state names from within the documents because state names
# were showing up in initial topics produced by STM
TextTokens <- tokens_remove(TextTokens, pattern = c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming', 'new', 'north', 'south', 'dakota', 'york', 'carolina', 'jersey', 'hampshire', 'mexico', 'rhode', 'island', 'west'))

# Compound key words and phrases that create necessary context
TextTokens <- tokens_compound(TextTokens, pattern = phrase(c("does not prohibit", "may prohibit", "shall prohibit", "not broadly prohibit", "not explicitly prohibit", "not specifically prohibit", "not require the removal", "require firearm removal", "does not require removal", "that require removal", "removal process", "exemption from removal requirement", "requires removal", "protective order", "restraining order", "temporary order", "temporary order", "order may", "order shall", "shall order", "may order", "law enforcement", "may petition for", "dating relationship", "child in common", "current or former", "domestic violence", "ex parte", "family or household member", "third party")))

# Now that tokens have been compounded to include key stopwords, 
# remove English stopwords from documents
ENstopwords <- stopwords(kind = "en")
TextTokens <- tokens_remove(TextTokens, pattern = ENstopwords)
TextTokens

Tokens consisting of 50 documents and 8 docvars.
alabama :
 [1] "law"               "domestic_violence" "firearm"          
 [4] "purchase"          "possession"        "prohibitions"     
 [7] "prohibits"         "following"         "individuals"      
[10] "owning"            "firearm"           "possessing"       
[ ... and 198 more ]

alaska :
 [1] "law"               "domestic_violence" "firearm"          
 [4] "purchase"          "possession"        "prohibitions"     
 [7] "may_prohibit"      "person"            "subject"          
[10] "protective_order"  "issued"            "notice"           
[ ... and 173 more ]

arizona :
 [1] "law"               "domestic_violence" "firearm"          
 [4] "purchase"          "possession"        "prohibitions"     
 [7] "prohibits"         "persons"           "convicted"        
[10] "misdemeanor"       "crime"             "domestic_violence"
[ ... and 198 more ]

arkansas :
 [1] "law"               "domestic_violence" "firearm"          
 [4] "purchase"          "possession"        "prohibitions"     
 [7] "does_not_prohibit" "following"         "persons"          
[10] "purchasing"        "possessing"        "firearms"         
[ ... and 85 more ]

california :
 [1] "law"               "domestic_violence" "firearm"          
 [4] "purchase"          "possession"        "prohibitions"     
 [7] "prohibits"         "following"         "purchasing"       
[10] "possessing"        "firearms"          "ammunition"       
[ ... and 757 more ]

colorado :
 [1] "law"               "domestic_violence" "firearm"          
 [4] "purchase"          "possession"        "prohibitions"     
 [7] "prohibits"         "following"         "persons"          
[10] "purchasing"        "possessing"        "firearms"         
[ ... and 443 more ]

[ reached max_ndoc ... 44 more documents ]

# Create document-feature matrix (DFM)
dfm <- dfm(TextTokens)

# Trim DFM, remove terms that appear in over 95% of documents
dfm_trim <- dfm_trim(dfm, max_docfreq = 0.95, docfreq_type = "prop")
dfm_trim

Document-feature matrix of: 50 documents, 2,002 features (91.25% sparse) and 8 docvars.
            features
docs         prohibits following owning possessing control
  alabama            2         3      1          1       1
  alaska             0         0      0          1       0
  arizona            2         1      0          4       0
  arkansas           0         2      0          1       0
  california         1         4      1          4       3
  colorado           1         3      0          3       6
            features
docs         misdemeanor offense valid protection domestic
  alabama              2       1     2         10        1
  alaska               1       0     0          0        0
  arizona              1       0     0          8        0
  arkansas             2       0     0          7        0
  california           3       1     0          1        0
  colorado             3       0     0         16        0
[ reached max_ndoc ... 44 more documents, reached max_nfeat ... 1,992 more features ]

# Feature Co-Occurrence Matrix for DFM
fcm2 <- fcm(dfm_trim)
# Pull top features
fcm_feats2 <- names(topfeatures(fcm2, 26))
# Retain top features in fcm
fcm2 <- fcm_select(fcm2, pattern = fcm_feats2, selection = "keep")

#jpeg("Matrix3.jpg", width = 1100, height = 500)
textplot_network(fcm2, edge_color = "darkred", edge_alpha = .3, vertex_labelcolor = "black", vertex_color = "darkgrey", vertex_labelsize = 6)

#dev.off()

The documents vary significantly in the number of tokens, but more tokens does not necessarily mean more firearm restrictions (which is why topic modeling is more helpful than looking at term frequency, etc.)

# Plot number of tokens by document
Meta_corpus %>%
  summary %>%
  ggplot(aes(x = State, y = Tokens, group = 1)) + theme_light() +
    geom_line(aes(color='darkred'), show.legend = F) +
    geom_point() +
  labs(title = "Number of Tokens in each Document (State)") +
     theme(axis.text.x = element_text(angle = 90)) +
  xlab(label = NULL) +
  ylab(label = "Number of Tokens") 

Structural Topic Modeling

# Convert DFM to STM
stm <- convert(dfm_trim, to = "stm")

Choosing K - number of topics

# Statistical fit criterion for evaluating different K's
# Fit models for K = 4 through 25
K <- c(4:25)
fit <- searchK(stm$documents, stm$vocab, K = K, verbose = FALSE)

# Create table with K value and coherence/exclusivity scores
Kplot <- data.frame("K" = K, 
                   "Coherence" = unlist(fit$results$semcoh),
                   "Exclusivity" = unlist(fit$results$exclus))
# Reshape table
Kplot <- melt(Kplot, id=c("K"))

#Plot results of K statistical testing
ggplot(Kplot, aes(K, value, color = variable)) +
  geom_line(size = 1.5, show.legend = FALSE) +
  facet_wrap(~variable,scales = "free_y") +
  labs(x = "Number of topics K",
       title = "Statistical fit of models with different K")

I will fit models with greater exclusivity, then hone in on a few topics produced that are coherent and relevant to the research question.

# Fit model for K = 5 for comparison of coherence/exclusivity
model_5 <- stm(documents = stm$documents,
         vocab = stm$vocab, 
         K = 5,
         verbose = FALSE)
# Fit model for K = 12 to evaluate increased exclusivity
model_12 <- stm(documents = stm$documents,
         vocab = stm$vocab, 
         K = 12,
         verbose = FALSE)
# Fit model for K = 20
model_20 <- stm(documents = stm$documents,
         vocab = stm$vocab, 
         K = 20,
         verbose = FALSE)
# Fit model for K = 25
model_25 <- stm(documents = stm$documents,
         vocab = stm$vocab, 
         K = 25,
         verbose = FALSE)

labelTopics(model_5, n=5)

Topic 1 Top Words:
     Highest Prob: protection, person, respondent, child, violation 
     FREX: partner, class, relief, spouse, misdemeanor 
     Lift: deemed, deny, directives, forms, high-risk 
     Score: chapter, injunction, card, in-laws, does_not_require_removal 
Topic 2 Top Words:
     Highest Prob: protective_order, person, ammunition, family, shall 
     FREX: revolver, protective_order, protective, violence, permits 
     Lift: contacting, regard, visitation, 1-5, acquaintance 
     Score: revolver, pistol, permits, post-separation, injunction 
Topic 3 Top Words:
     Highest Prob: party, protection, law_enforcement, person, shall 
     FREX: adverse, party, abusing, restrained, party's 
     Lift: accessing, displayed, restraining_order_shall, scene, 13within 
     Score: adverse, pistol, concealed, obtaining, party's 
Topic 4 Top Words:
     Highest Prob: respondent, shall, surrender, person, protective_order 
     FREX: respondent, hearing, possessed, respondent's, injunction 
     Lift: 2nd, constituting, extend, satisfied, vulnerable 
     Score: injunction, obey, stay, 3rd, 2nd 
Topic 5 Top Words:
     Highest Prob: abuse, ammunition, defendant, weapons, shall 
     FREX: muzzle-loading, relinquished, complaint, weapons, defendant 
     Lift: 1or, 22within, accessed, accidental, acknowledgement 
     Score: rifles, shotguns, muzzle-loading, acknowledgment, large 

plot(model_5, type = "summary", labeltype = "frex", text.cex = 0.7, main = "Topic Shares Across the Corpus", xlab = "Estimated Share of Topic")

As expected, 5 topics is not exclusive enough for my analysis

labelTopics(model_12, n=5)

Topic 1 Top Words:
     Highest Prob: protection, ammunition, violation, may_petition_for, final 
     FREX: relief, misdemeanor, does_not_prohibit, necessary, related 
     Lift: allowed, directives, a-g, i-iv, in-laws 
     Score: does_not_prohibit, does_not_require_removal, ex_parte, relief, adoption 
Topic 2 Top Words:
     Highest Prob: protective_order, family, person, abuse, violence 
     FREX: violence, protective, family, transferred, protective_order 
     Lift: regard, benefit, contracted, convictions, defamation 
     Score: injunction, post-separation, protective_order, divorce, transferred 
Topic 3 Top Words:
     Highest Prob: party, protection, abusing, law_enforcement, agency 
     FREX: adverse, party, abusing, restrained, party's 
     Lift: 13within, 3and, acknowledge, adjudications, adverse 
     Score: adverse, restrained, party, abusing, licensee 
Topic 4 Top Words:
     Highest Prob: respondent, surrender, person, protective_order, shall 
     FREX: injunction, hearing, stay, eligible, adult 
     Lift: vulnerable, 9a, abode, amendment, appeared 
     Score: injunction, stay, sheriff, protective_order, indicates 
Topic 5 Top Words:
     Highest Prob: abuse, domestic, defendant, person, protection 
     FREX: domestic, guns, rifles, shotguns, large 
     Lift: borrow, departmental, devices, enables, feeding 
     Score: rifles, shotguns, abuse, card, large 
Topic 6 Top Words:
     Highest Prob: defendant, weapon, ammunition, deadly, weapons 
     FREX: located, deadly, purchaser, defendant, identification 
     Lift: accompany, facsimile, guard, proceed, restraining_order_may 
     Score: defendant, purchaser, located, card, belonging 
Topic 7 Top Words:
     Highest Prob: protection, respondent, shall, use, degree 
     FREX: temporary_order, obey, 3rd, possessed, 2nd 
     Lift: 2nd, constituting, 3rd, bail, deliberate 
     Score: obey, 3rd, 2nd, constituting, rifles 
Topic 8 Top Words:
     Highest Prob: pistol, person, dangerous, surrender, weapons 
     FREX: revolver, pistol, concealed, dangerous, obtaining 
     Lift: accessing, certificate, acceptance, allows, amount 
     Score: pistol, revolver, obtaining, concealed, license 
Topic 9 Top Words:
     Highest Prob: ammunition, weapons, defendant, abuse, shall 
     FREX: permits, safekeeping, third_party, aggrieved, relinquished 
     Lift: accepts, acknowledgement, armory, birth, caption 
     Score: sheriff, permits, third_party, acknowledgment, defendant 
Topic 10 Top Words:
     Highest Prob: person, protection, respondent, child, abuse 
     FREX: muzzle-loading, partner, intimate, card, qualified 
     Lift: high-risk, plenary, preceding, abstain, aged 
     Score: muzzle-loading, card, partner, bow, bows 
Topic 11 Top Words:
     Highest Prob: protection, ammunition, person, personal, injunction 
     FREX: shipping, personal, transporting, receiving, injunction 
     Lift: abusive, accorded, activity, affection, alternatively 
     Score: injunction, shipping, transporting, personal, specific 
Topic 12 Top Words:
     Highest Prob: respondent, protection, ammunition, spouse, weapon 
     FREX: chapter, date, spouse, respondent's, check 
     Lift: forms, 12.1-16, 12.1-17, 12.1-17.2, 12.1-18 
     Score: chapter, private, check, spouse, bureau 

# Plot topic shares across corpus for model with 12 topics - FREX weight applied to labels
plot(model_12, type = "summary", labeltype = "frex", text.cex = 0.7, main = "Topic Shares Across the Corpus", xlab = "Estimated Share of Topic")

labelTopics(model_20, n=5)

Topic 1 Top Words:
     Highest Prob: protection, may_petition_for, misdemeanor, respondent, issuing 
     FREX: together, relief, misdemeanor, resided, may_petition_for 
     Lift: imprisonment8, in-laws, misdemeanor7, parent-in-law, specifically 
     Score: protection, resided, together, does_not_require_removal, does_not_prohibit 
Topic 2 Top Words:
     Highest Prob: violence, ammunition, injunction, person, family 
     FREX: violence, injunction, dating, punished, minor 
     Lift: punished, 1-3, 2,000-, 3,000-, accorded 
     Score: injunction, violence, punished, bring, dating 
Topic 3 Top Words:
     Highest Prob: party, abusing, adverse, transfer, protection 
     FREX: adverse, abusing, party's, party, transfer 
     Lift: 3and, acknowledge, adjudications, agreeing, appointed 
     Score: adverse, abusing, party, party's, transferred 
Topic 4 Top Words:
     Highest Prob: protective_order, respondent, person, law_enforcement, may 
     FREX: eligible, protective_order, relief, vulnerable, minors 
     Lift: cal, fam, qualify, regulated, stolen 
     Score: protective_order, respondent, law_enforcement, vulnerable, eligible 
Topic 5 Top Words:
     Highest Prob: domestic, abuse, defendant, person, protective_order 
     FREX: domestic, large, guns, rifles, shotguns 
     Lift: devices, non-large, 1or, accidental, administrative 
     Score: domestic, rifles, shotguns, guns, card 
Topic 6 Top Words:
     Highest Prob: defendant, ammunition, weapons, abuse, shall 
     FREX: safekeeping, defendant's, acknowledgment, relinquished, third_party 
     Lift: accepts, acknowledgement, armory, arranging, birth 
     Score: defendant, third_party, sheriff, relinquished, weapons 
Topic 7 Top Words:
     Highest Prob: protection, respondent, shall, use, assault 
     FREX: temporary_order, obey, 3rd, possessed, rifles 
     Lift: 2nd, 3rd, bail, constituting, deliberate 
     Score: obey, 3rd, rifles, shotguns, temporary_order 
Topic 8 Top Words:
     Highest Prob: pistol, person, dangerous, weapons, surrender 
     FREX: revolver, pistol, concealed, dangerous, obtaining 
     Lift: accessing, acceptance, allows, amount, antiharassment 
     Score: pistol, revolver, concealed, dangerous, license 
Topic 9 Top Words:
     Highest Prob: ammunition, protective_order, shall, respondent, permits 
     FREX: permits, aggrieved, qualified, step, protective_order 
     Lift: academy, agent, aggrieved, annulment, appeals 
     Score: permits, protective_order, aggrieved, qualified, step 
Topic 10 Top Words:
     Highest Prob: person, protection, defendant, child, abuse 
     FREX: muzzle-loading, partner, intimate, card, child 
     Lift: abstain, bow, bows, crossbow, crossbows 
     Score: muzzle-loading, card, partner, defendant, bow 
Topic 11 Top Words:
     Highest Prob: protection, restrained, party, law_enforcement, member 
     FREX: restrained, licensee, party, receipt, federal 
     Lift: 13within, activity, advisement, arising, charging 
     Score: restrained, licensee, party, law_enforcement, incident 
Topic 12 Top Words:
     Highest Prob: protection, respondent, person, violation, spouse 
     FREX: chapter, spouse, class, offenses, parent 
     Lift: forms, 12.1-16, 12.1-17, 12.1-17.2, 12.1-18 
     Score: chapter, respondent, dangerous, natural, class 
Topic 13 Top Words:
     Highest Prob: weapon, defendant, person, purchaser, shall 
     FREX: purchaser, identification, located, belonging, duty 
     Lift: accompany, anticipates, antique, appropriately, assignee 
     Score: purchaser, identification, card, located, belonging 
Topic 14 Top Words:
     Highest Prob: respondent, injunction, surrender, person, shall 
     FREX: injunction, stay, form, s, sheriff 
     Lift: 9a, abode, appeared, appears, approves 
     Score: injunction, sheriff, respondent, stay, third_party 
Topic 15 Top Words:
     Highest Prob: person, ammunition, family_or_household_member, abuse, defined 
     FREX: family_or_household_member, intentionally, police, restraining, petitioner 
     Lift: administer, adopt, anyone, beneficiaries, bound 
     Score: restraining_order, restraining, police, qualifying, family_or_household_member 
Topic 16 Top Words:
     Highest Prob: ammunition, protection, possessing, person, receiving 
     FREX: shipping, transporting, commit, receiving, territory 
     Lift: apparent, as-applied, clause, concluded, creating 
     Score: shipping, commit, transporting, findings, same-sex 
Topic 17 Top Words:
     Highest Prob: respondent, ammunition, protection, abuse, law_enforcement 
     FREX: prevention, federally, dealer, licensed, relinquish 
     Lift: bureau, group, accessed, acknowledges, adequately 
     Score: prevention, dealer, federally, respondent, abuse 
Topic 18 Top Words:
     Highest Prob: ammunition, deadly, weapons, protective_order, specified 
     FREX: deadly, specified, ownership, peace, duration 
     Lift: a-g, allowed, arraignment, arrests, consummated 
     Score: deadly, specified, protective_order, weapons, ownership 
Topic 19 Top Words:
     Highest Prob: protective_order, family, person, abuse, violence 
     FREX: divorce, transferred, family, post-separation, protective 
     Lift: benefit, hard, labor, parish, places 
     Score: protective_order, injunction, sheriff, transferred, violence 
Topic 20 Top Words:
     Highest Prob: personal, protection, respondent, individual, contempt 
     FREX: personal, contempt, abusive, alleged, motion 
     Lift: abusive, alternatively, answer, arrested, bench 
     Score: personal, abusive, answer, enjoin, imposes 

# Plot topic shares across corpus for model with 17 topics - highest word probability as labels
plot(model_20, type = "summary", text.cex = 0.7, main = "Topic Shares Across the Corpus", xlab = "Estimated Share of Topic")

labelTopics(model_25, n=5)

Topic 1 Top Words:
     Highest Prob: protection, may_petition_for, misdemeanor, purchasing, following 
     FREX: deems, purchasing, does_not_prohibit, acts, relief 
     Lift: in-laws, find, cohabitants, 4th, deems 
     Score: protection, deems, does_not_prohibit, purchasing, acts 
Topic 2 Top Words:
     Highest Prob: violence, injunction, ammunition, person, minor 
     FREX: violence, injunction, dating, punished, committing 
     Lift: punished, 1-3, 2,000-, 3,000-, accorded 
     Score: injunction, violence, punished, bring, protective_order 
Topic 3 Top Words:
     Highest Prob: adverse, party, control, protection, surrender 
     FREX: adverse, party's, extended, party, control 
     Lift: adverse, appointed, category, concerning, informing 
     Score: adverse, extended, party's, party, control 
Topic 4 Top Words:
     Highest Prob: respondent, protective_order, surrender, relief, person 
     FREX: eligible, vulnerable, minors, adults, relief 
     Lift: vulnerable, authorities, barracks, both12, classified 
     Score: vulnerable, protective_order, respondent, minors, adults 
Topic 5 Top Words:
     Highest Prob: domestic, abuse, protective_order, defendant, person 
     FREX: domestic, complaint, suffering, receive, minors 
     Lift: rights, 1or, accidental, administrative, amilies 
     Score: domestic, protective_order, abuse, defendant, suffering 
Topic 6 Top Words:
     Highest Prob: person, defendant, protection, child, partner 
     FREX: qualified, partner, grandparent, individual, defendant 
     Lift: arrangement, e, grandparent-in-law, imprisonment8, maintained 
     Score: defendant, qualified, grandparent, partner, transfer 
Topic 7 Top Words:
     Highest Prob: protection, respondent, shall, use, degree 
     FREX: obey, 3rd, temporary_order, rifles, shotguns 
     Lift: 2nd, constituting, 16the, 1st, 3rd 
     Score: obey, 3rd, temporary_order, 2nd, constituting 
Topic 8 Top Words:
     Highest Prob: pistol, person, dangerous, weapons, surrender 
     FREX: revolver, pistol, concealed, dangerous, obtaining 
     Lift: accessing, acceptance, allows, amount, antiharassment 
     Score: pistol, revolver, dangerous, concealed, license 
Topic 9 Top Words:
     Highest Prob: ammunition, protective_order, respondent, shall, permits 
     FREX: permits, aggrieved, step, qualified, protective_order 
     Lift: step, subdivision, academy, agent, aggrieved 
     Score: permits, protective_order, aggrieved, qualified, step 
Topic 10 Top Words:
     Highest Prob: protection, person, abuse, respondent, defendant 
     FREX: muzzle-loading, card, partner, intimate, bow 
     Lift: bow, bows, crossbow, crossbows, high-risk 
     Score: muzzle-loading, card, bow, bows, crossbow 
Topic 11 Top Words:
     Highest Prob: protection, abuse, household, member, domestic 
     FREX: living, household, member, carry, license 
     Lift: activity, defense, menace, quarters, self 
     Score: living, abuse, member, domestic, protection 
Topic 12 Top Words:
     Highest Prob: protection, respondent, violation, spouse, final 
     FREX: chapter, spouse, class, dangerous, committing 
     Lift: forms, natural, 12.1-16, 12.1-17, 12.1-17.2 
     Score: chapter, respondent, protection, spouse, dangerous 
Topic 13 Top Words:
     Highest Prob: weapon, defendant, person, purchaser, shall 
     FREX: purchaser, identification, located, belonging, duty 
     Lift: accompany, anticipates, antique, appropriately, assignee 
     Score: purchaser, located, identification, card, belonging 
Topic 14 Top Words:
     Highest Prob: respondent, injunction, surrender, person, hearing 
     FREX: injunction, stay, form, s, sheriff 
     Lift: 9a, abode, appeared, appears, approves 
     Score: injunction, sheriff, stay, respondent, s 
Topic 15 Top Words:
     Highest Prob: person, child, abuse, family_or_household_member, defined 
     FREX: unmarried, family_or_household_member, intimate, petitioner, cohabited 
     Lift: preceding, anyone, element, estrains, intimidating 
     Score: unmarried, family_or_household_member, considers, qualifying, abuse 
Topic 16 Top Words:
     Highest Prob: ammunition, protection, possessing, person, receiving 
     FREX: shipping, receiving, commit, transporting, territory 
     Lift: apparent, as-applied, clause, concluded, creating 
     Score: shipping, transporting, findings, same-sex, sentencing 
Topic 17 Top Words:
     Highest Prob: respondent, ammunition, protection, abuse, law_enforcement 
     FREX: prevention, federally, dealer, relinquish, licensed 
     Lift: bureau, group, accessed, acknowledges, adequately 
     Score: prevention, dealer, federally, respondent, abuse 
Topic 18 Top Words:
     Highest Prob: protection, ammunition, notice, person, respondent 
     FREX: notice, deadly, may_order, currently, subjects 
     Lift: a-g, allowed, i-iv, invasion, pon 
     Score: allowed, protection, directives, deadly, chapter 
Topic 19 Top Words:
     Highest Prob: protective_order, family, person, violence, abuse 
     FREX: post-separation, divorce, cases, violence, transferred 
     Lift: advise, affectionate, agreed, applicant's, asks 
     Score: protective_order, injunction, sheriff, transferred, post-separation 
Topic 20 Top Words:
     Highest Prob: protection, respondent, assault, family, victim 
     FREX: assault, knowingly, transporting, months, victim 
     Lift: order_may_prohibit, post-release, violate, supervision, transport 
     Score: transporting, respondent, knowingly, protection, assault 
Topic 21 Top Words:
     Highest Prob: protective_order, respondent, may, law_enforcement, person 
     FREX: protective_order, determination, gun, cohabitant, dating 
     Lift: cal, fam, qualify, stolen, amendment 
     Score: protective_order, determination, respondent, cal, fam 
Topic 22 Top Words:
     Highest Prob: ammunition, defendant, deadly, weapons, protective_order 
     FREX: deadly, ownership, specified, seize, weapons 
     Lift: beneficiaries, disqualification, facsimile, indictment, reciprocal 
     Score: deadly, defendant, ownership, weapons, protective_order 
Topic 23 Top Words:
     Highest Prob: personal, protection, respondent, individual, contempt 
     FREX: personal, motion, arrest, abusive, contempt 
     Lift: abusive, alternatively, answer, arrested, bench 
     Score: personal, abusive, answer, enjoin, imposes 
Topic 24 Top Words:
     Highest Prob: ammunition, abuse, defendant, weapons, protection 
     FREX: safekeeping, rifles, shotguns, guns, large 
     Lift: accepts, acknowledgement, armory, birth, caption 
     Score: third_party, defendant, sheriff, weapons, abuse 
Topic 25 Top Words:
     Highest Prob: party, protection, abusing, restrained, law_enforcement 
     FREX: restrained, abusing, party, licensee, federally 
     Lift: 13within, 3and, acknowledge, adjudications, advisement 
     Score: restrained, abusing, party, licensee, transfer 

# Plot topic shares across corpus for model with 25 topics - highest word probability as labels
plot(model_25, type = "summary", text.cex = 0.7, main = "Topic Shares Across the Corpus", xlab = "Estimated Share of Topic")

K = 20 seems to provide decently exclusive topics with a few key topics that I can use for analysis.

It seems the “score” weight produces some pretty coherent topic features. I’ll pull out a few of the topics that seem to cover key interests of this study.

• Topic 1 will be my primary focus because it includes the phrases “does not require removal” and “does not prohibit”. This will allow me to identify documents that these phrases are prevalent in, implying limited firearm prohibitions and provisions.

• Topic 4 includes eligibility for restraining orders.

• Topic 15 covers the nature of relationship between a victim and abuser. Each state defines intimate partners differently and which types of relationships are protected under firearm provisions.

• Topics 6, 17, and 18 seem similar, so I’ll evaluate which best covers the topic of weapon ownership and relinquishment for the defendant.

# Plot topics 6 and 17 to compare
plot(model_20, type = "perspectives", topics = c(6,17), main = "Topic contrasts", text.cex = .8)

# Plot topics 18 and 17 to compare
plot(model_20, type = "perspectives", topics = c(18,17), main = "Topic contrasts", text.cex = .8)

It looks like Topic 6 will best cover weapon relinquishment.

# Select topics 1, 4, 6, 15 due to relevance of the research question
#jpeg("TopNamesNew.jpg", width = 1400, height = 1000, pointsize = 35)
plot(model_20, type = "labels", topics = c(1, 4, 6, 15), labeltype = "score", main = "Top Words (weighted by score) for Topics 1, 4, 6, 15", text.cex = .85, n=10, width = 68)

#dev.off()

# Table of top 10 words (Score weight) for stm model (K=20) 
topics_graph <- labelTopics(model_20, n=10)
topics_graph <- data.frame("features" = t(topics_graph$score))
colnames(topics_graph) <- paste("Topic", c(1:20))
# Select relevant topics (1, 4, 6, 15)
select(topics_graph, "Topic 1", "Topic 4", "Topic 6", "Topic 15") %>%
   kable() %>% kable_styling() %>% add_header_above(c("Top 10 Features (Weighted by Score) for Selected Topics"=4)) 

Top 10 Features (Weighted by Score) for Selected Topics
Topic 1	Topic 4	Topic 6	Topic 15
protection	protective_order	defendant	restraining_order
resided	respondent	third_party	restraining
together	law_enforcement	sheriff	police
does_not_require_removal	vulnerable	relinquished	qualifying
does_not_prohibit	eligible	weapons	family_or_household_member
respondent	determination	acknowledgment	chief
relief	minors	safekeeping	considers
currently	dating	abuse	intentionally
knowingly	regulated	defendant’s	beneficiaries
reside	cal	relinquish	disqualification

# Plot distribution of top 10 words from each topic 
Beta <- tidy(model_20, matrix = "beta")
Beta <- Beta %>% filter(topic == "1" | topic == "4" | topic == "6" | topic == "15")
Beta_terms <-  Beta %>%
  group_by(topic) %>%
  slice_max(beta, n = 10) %>% 
  ungroup() %>%
  arrange(topic, -beta)
Beta_terms %>%
  mutate(term = reorder_within(term, beta, topic)) %>%
  ggplot(aes(beta, term, fill = factor(topic))) +
  geom_col(show.legend = FALSE) +
  labs(title = "Distribution Witin Topics of Top 10 features (beta)") +
  facet_wrap(~ topic, scales = "free") +
  scale_y_reordered()

Based on the topic correlation plot below, it looks like the topics I selected are pretty unique from each other, which will help avoid overly redundant analysis.

# Plot topic correlation
mod.out.corr <- topicCorr(model_20)
plot(mod.out.corr, vertex.size = 30, vertex.color = "#FFCCCC", vertex.label.cex = 1.1, vlabels = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20))

# make a document-topic-matrix looking at topic prevalence across the states
# topic association by state (gamma)
gamma <- tidy(model_20, matrix = "gamma") %>%
  pivot_wider(names_from = topic, values_from = gamma)
gamma$state <- c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming')
gamma <- gamma[, c('document', 'state', '1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20')]
gamma

# A tibble: 50 x 22
   document state          `1`     `2`     `3`     `4`     `5`     `6`
      <int> <chr>        <dbl>   <dbl>   <dbl>   <dbl>   <dbl>   <dbl>
 1        1 alabama    3.04e-1 8.56e-4 9.94e-4 6.83e-4 6.42e-4 3.79e-4
 2        2 alaska     1.73e-1 1.89e-3 9.36e-4 7.87e-1 6.59e-4 4.77e-4
 3        3 arizona    6.15e-1 7.66e-4 2.33e-2 1.13e-3 1.86e-3 6.21e-2
 4        4 arkansas   9.72e-1 5.06e-3 1.95e-3 7.60e-4 1.43e-3 5.06e-5
 5        5 california 1.26e-3 7.65e-5 3.24e-5 9.96e-1 5.49e-5 7.10e-5
 6        6 colorado   6.16e-4 4.98e-5 1.44e-4 3.38e-4 2.15e-4 5.52e-5
 7        7 connectic~ 6.63e-4 5.58e-5 1.01e-4 2.21e-4 1.63e-4 7.61e-5
 8        8 delaware   1.71e-3 5.94e-5 1.49e-4 6.63e-4 1.85e-4 2.38e-4
 9        9 florida    1.55e-1 8.40e-1 4.77e-4 3.98e-4 1.42e-4 1.82e-6
10       10 georgia    2.74e-1 7.11e-1 9.40e-4 1.22e-3 3.81e-4 7.48e-6
# ... with 40 more rows, and 14 more variables: `7` <dbl>, `8` <dbl>,
#   `9` <dbl>, `10` <dbl>, `11` <dbl>, `12` <dbl>, `13` <dbl>,
#   `14` <dbl>, `15` <dbl>, `16` <dbl>, `17` <dbl>, `18` <dbl>,
#   `19` <dbl>, `20` <dbl>

# Plot topic association (gamma) for each state
gamma1 <- gamma %>%
  pivot_longer(c('1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20'), names_to = "topic", values_to = "gamma") 
gamma1 %>% 
  ggplot(aes(factor(topic), gamma)) +
  geom_boxplot() +
  facet_wrap(~ state) +
  labs(x = "topic", y = expression(gamma))

Now that I’ve selected a model to use and the specific topics of interest, I will proceed with some analysis. I want to look at how the 4 selected topics differ in proportion/prevalence across each document (state) and how the presence or omission of topics relate to homicide rates.

# DTM with focus on topics 1, 4, 6, 15
# topic association by state (gamma)
gammaX <- tidy(model_20, matrix = "gamma") %>%
  pivot_wider(names_from = topic, values_from = gamma)
gammaX$state <- c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming')
gammaX <- gammaX[, c('document', 'state', '1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20')]
gammaX <- select(gammaX, 'document', 'state', '1', '4', '6', '15')

# Plot topic association (gamma) for each state
gammaX <- gammaX %>%
  pivot_longer(c('1', '4', '6', '15'), names_to = "topic", values_to = "gamma") 
gammaX %>% 
  ggplot(aes(factor(topic), gamma)) +
  geom_boxplot() +
  facet_wrap(~ state) +
  labs(x = "Topic", y = expression(gamma))

Plot 1: Topic Proportion and Female Homicide Rates

# Evaluate covariate female homicide rates in relation to topic proportion
# Plot 1: Topic Proportion and Female Homicide Rates

model20_labels <- c("Topic 1", "Topic 2", "Topic 3", "Topic 4", "Topic 5", "Topic 6", "Topic 7", "Topic 8", "Topic 9", "Topic 10", "Topic 11", "Topic 12", "Topic 13", "Topic 14", "Topic 15", "Topic 16", "Topic 17", "Topic 18", "Topic 19", "Topic 20")

MetaData$Crude.Rate_Female <- as.numeric(MetaData$Crude.Rate_Female)
prep <- estimateEffect(1:20 ~ Crude.Rate_Female, model_20, meta=MetaData, 
                       uncertainty="Global")

#jpeg("Top1Prop.jpg", width = 1500, height = 1500, pointsize = 50)
par(mfrow=c(2,2))
for (i in c(1,4,6,15)){
  plot(prep, "Crude.Rate_Female", method = "continuous", topics = i, main = paste0(model20_labels[i]), ylab = "Topic Proportion", xlab = "Homicide Rate (Female)", printlegend = F, ps=20, linecol = "darkred")}

#dev.off()

# Same as above, but just for Topic 1 for larger view
#jpeg("Prop1.jpg", width = 1800, height = 1400, pointsize = 50)

plot(prep, "Crude.Rate_Female", method = "continuous", topics = 1, main = "Topic 1", ylab = "Topic Proportion", xlab = "Homicide Rate (Female)", printlegend = F, ps=20, linecol = "darkred")

#dev.off()

Now I will plot the topic prevalence for Topic 1 for the 10 states with the highest female homicide rates and the 10 states with the lowest female homicide rates.

# Find the states with the highest and lowest female homicide rates
FemaleHomRates <- CDCHomRates %>% arrange(desc(Crude.Rate_Female))
#head(FemaleHomRates, 10)
#tail(FemaleHomRates, 10)

# Make a document-topic-matrix topic association by state (gamma)
gamma <- tidy(model_20, matrix = "gamma") %>%
  pivot_wider(names_from = topic, values_from = gamma)
gamma$document <- c('alabama', 'alaska', 'arizona', 'arkansas', 'california', 'colorado', 'connecticut', 'delaware', 'florida', 'georgia', 'hawaii', 'idaho', 'illinois', 'indiana', 'iowa', 'kansas', 'kentucky', 'louisiana', 'maine', 'maryland', 'massachusetts', 'michigan', 'minnesota', 'mississippi', 'missouri', 'montana', 'nebraska', 'nevada', 'new-hampshire', 'new-jersey', 'new-mexico', 'new-york', 'north-carolina', 'north-dakota', 'ohio', 'oklahoma', 'oregon', 'pennsylvania', 'rhode-island', 'south-carolina', 'south-dakota', 'tennessee', 'texas', 'utah', 'vermont', 'virginia', 'washington', 'west-virginia', 'wisconsin', 'wyoming')
gamma1 <- select(gamma, document, '1') 

gammaHigh <- gamma1 %>%
  filter(document == 'mississippi' | document == 'alabama' | document == 'alaska' | document == 'missouri' | document == 'louisiana' | document == 'arkansas' | document == 'new-mexico' | document == 'south-carolina' | document == 'tennessee' | document == 'nevada')
gammaHigh <- as.data.frame(gammaHigh)
gammaHigh$Hom_Rates <- c("High", "High", "High", "High", "High", "High", "High", "High", "High", "High")

gammaLow <- gamma1 %>%
 filter(document == 'oregon' | document == 'connecticut' | document == 'iowa' | document == 'minnesota' | document == 'new-york' | document == 'massachusetts' | document == 'north-dakota' | document == 'rhode-island' | document == 'hawaii' | document == 'south-dakota')
gammaLow <- as.data.frame(gammaLow)
gammaLow$Hom_Rates <- c("Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low")

#Reorder states in High and Low category from lowest homicide rate to highest
LowStates <- c("south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")
gammaLowNew <- gammaLow[match(LowStates, gammaLow$document), ]
HighStates <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi")
gammaHighNew <- gammaHigh[match(HighStates, gammaHigh$document), ]

gamma_LowHigh <- rbind(gammaHighNew, gammaLowNew)
gamma_LowHigh <- rename(gamma_LowHigh, "gamma_topic_1" = "1")

Vec <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi", "south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")

Top1Prev <- ggplot(data = gamma_LowHigh, mapping = aes(x = factor(document, levels = unique(Vec)), y = gamma_topic_1)) + 
  geom_point(color = "darkred", size = 3) +
  facet_wrap(vars(Hom_Rates), scales = "free_x") +
  labs(title = "Plot 2: Prevalence of Topic 1", 
       subtitle = "for States with Highest and Lowest Female Homicide Rates", x = "State", y = "Prevalence of Topic 1 (gamma)") +
  theme_light(base_size = 10) +
  theme(axis.text.x = element_text(angle = 90))
#ggsave("Top1Prev.jpg", width = 5, height = 3)

Top1Prev

# Plot the prevalence of Topic 1 in relation to confirmed intimate partner homicide rates
gammaNV <- tidy(model_20, matrix = "gamma") %>%
  pivot_wider(names_from = topic, values_from = gamma)

gammaNV$document <- c("Alabama",  "Alaska", "Arizona", "Arkansas", "California",     "Colorado",       "Connecticut",    "Delaware",      
 "Florida",        "Georgia",        "Hawaii",        "Idaho",         
 "Illinois",       "Indiana",        "Iowa",           "Kansas",        
 "Kentucky",       "Louisiana",      "Maine",          "Maryland",      
 "Massachusetts",  "Michigan",       "Minnesota",      "Mississippi",   
 "Missouri",       "Montana",        "Nebraska",       "Nevada",       
 "New Hampshire",  "New Jersey",     "New Mexico",     "New York",      
 "North Carolina", "North Dakota",   "Ohio",           "Oklahoma",      
 "Oregon",         "Pennsylvania",   "Rhode Island",   "South Carolina",
 "South Dakota",   "Tennessee",      "Texas",          "Utah",          
 "Vermont",        "Virginia",       "Washington",     "West Virginia", 
 "Wisconsin",      "Wyoming")

gammaNV <- filter(gammaNV, document == "Alaska" | document == "Arizona" | document == "California" | document == "Colorado" | document == "Connecticut" | document == "Delaware" | document == "Georgia" | document == "Illinois" |document == "Indiana" |document == "Iowa" | document == "Kansas" | document == "Kentucky" | document == "Maine" | document == "Maryland" | document == "Massachusetts" | document == "Michigan" | document == "Minnesota" | document == "Nevada" | document == "New Hampshire" | document == "New Jersey" | document == "New Mexico" | document == "North Carolina" | document == "Ohio" | document == "Oklahoma" | document == "Oregon" | document == "Pennsylvania" | document == "South Carolina" | document == "Utah" | document == "Virginia" | document == "Washington" | document == "West Virginia" | document == "Wisconsin")
gammaNV <- select(gammaNV, document, '1')

# | document == "Rhode Island" , document == "Vermont" | 
# Omit suppressed values to allow for plotting 

NV <- NVDRSdata %>% 
  filter(!str_detect(State, "Rhode Island")) %>%
  filter(!str_detect(State, "Vermont"))
gammaNV$HomRate <- NV$Crude_Rate
gammaNV <- rename(gammaNV, "GammaTop1" = "1")
gammaNV$HomRate <- as.numeric(gammaNV$HomRate)

IPHgamma <- ggplot(data = gammaNV, mapping = aes(x = HomRate, y = GammaTop1)) + 
  geom_line(aes(color='darkred'), show.legend = F) +
    geom_point(show.legend = F) + theme_light(base_size = 10) +
  geom_text_repel(aes(label=ifelse(GammaTop1>.05, as.character(document),'')), angle = 0, hjust = -.1, vjust = .3, size = 3.5, point.padding = -1, box.padding = -.3) +
  labs(title = "Plot 3: Topic 1 Prevalence and Intimate Partner Homicide Rates", 
       subtitle = "for States with Available Data", x = "Intimate Partner Homicide Rate", y = "Prevalence of Topic 1 (gamma)")  +
  theme(axis.text.x = element_text(angle = 0))
#ggsave("IPHgamma1.jpg", width = 5, height = 3)

IPHgamma

gamma4 <- select(gamma, document, '4') 
gammaHigh <- gamma4 %>%
  filter(document == 'mississippi' | document == 'alabama' | document == 'alaska' | document == 'missouri' | document == 'louisiana' | document == 'arkansas' | document == 'new-mexico' | document == 'south-carolina' | document == 'tennessee' | document == 'nevada')
gammaHigh <- as.data.frame(gammaHigh)
gammaHigh$Hom_Rates <- c("High", "High", "High", "High", "High", "High", "High", "High", "High", "High")
gammaLow <- gamma4 %>%
 filter(document == 'oregon' | document == 'connecticut' | document == 'iowa' | document == 'minnesota' | document == 'new-york' | document == 'massachusetts' | document == 'north-dakota' | document == 'rhode-island' | document == 'hawaii' | document == 'south-dakota')
gammaLow <- as.data.frame(gammaLow)
gammaLow$Hom_Rates <- c("Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low")
LowStates <- c("south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")
gammaLowNew <- gammaLow[match(LowStates, gammaLow$document), ]
HighStates <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi")
gammaHighNew <- gammaHigh[match(HighStates, gammaHigh$document), ]
gamma_LowHigh <- rbind(gammaHighNew, gammaLowNew)
gamma_LowHigh <- rename(gamma_LowHigh, "gamma_topic_4" = "4")

Vec <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi", "south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")

Top4Prev <- ggplot(data = gamma_LowHigh, mapping = aes(x = factor(document, levels = unique(Vec)), y = gamma_topic_4)) + 
  geom_point(color = "darkred", size = 3) +
  facet_wrap(vars(Hom_Rates), scales = "free_x") +
  labs(title = "Prevalence of Topic 4", 
       subtitle = "for States with Highest and Lowest Female Homicide Rates", x = "State", y = "Prevalence of Topic 4 (gamma)") +
  theme_light(base_size = 10) +
  theme(axis.text.x = element_text(angle = 90))
Top4Prev

gamma6 <- select(gamma, document, '6') 
gammaHigh <- gamma6 %>%
  filter(document == 'mississippi' | document == 'alabama' | document == 'alaska' | document == 'missouri' | document == 'louisiana' | document == 'arkansas' | document == 'new-mexico' | document == 'south-carolina' | document == 'tennessee' | document == 'nevada')
gammaHigh <- as.data.frame(gammaHigh)
gammaHigh$Hom_Rates <- c("High", "High", "High", "High", "High", "High", "High", "High", "High", "High")
gammaLow <- gamma6 %>%
 filter(document == 'oregon' | document == 'connecticut' | document == 'iowa' | document == 'minnesota' | document == 'new-york' | document == 'massachusetts' | document == 'north-dakota' | document == 'rhode-island' | document == 'hawaii' | document == 'south-dakota')
gammaLow <- as.data.frame(gammaLow)
gammaLow$Hom_Rates <- c("Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low")
LowStates <- c("south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")
gammaLowNew <- gammaLow[match(LowStates, gammaLow$document), ]
HighStates <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi")
gammaHighNew <- gammaHigh[match(HighStates, gammaHigh$document), ]
gamma_LowHigh <- rbind(gammaHighNew, gammaLowNew)
gamma_LowHigh <- rename(gamma_LowHigh, "gamma_topic_6" = "6")

Vec <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi", "south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")

Top6Prev <- ggplot(data = gamma_LowHigh, mapping = aes(x = factor(document, levels = unique(Vec)), y = gamma_topic_6)) + 
  geom_point(color = "darkred", size = 3) +
  facet_wrap(vars(Hom_Rates), scales = "free_x") +
  labs(title = "Prevalence of Topic 6", 
       subtitle = "for States with Highest and Lowest Female Homicide Rates", x = "State", y = "Prevalence of Topic 6 (gamma)") +
  theme_light(base_size = 10) +
  theme(axis.text.x = element_text(angle = 90))
Top6Prev

gamma15 <- select(gamma, document, '15') 
gammaHigh <- gamma15 %>%
  filter(document == 'mississippi' | document == 'alabama' | document == 'alaska' | document == 'missouri' | document == 'louisiana' | document == 'arkansas' | document == 'new-mexico' | document == 'south-carolina' | document == 'tennessee' | document == 'nevada')
gammaHigh <- as.data.frame(gammaHigh)
gammaHigh$Hom_Rates <- c("High", "High", "High", "High", "High", "High", "High", "High", "High", "High")
gammaLow <- gamma15 %>%
 filter(document == 'oregon' | document == 'connecticut' | document == 'iowa' | document == 'minnesota' | document == 'new-york' | document == 'massachusetts' | document == 'north-dakota' | document == 'rhode-island' | document == 'hawaii' | document == 'south-dakota')
gammaLow <- as.data.frame(gammaLow)
gammaLow$Hom_Rates <- c("Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low", "Low")
LowStates <- c("south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")
gammaLowNew <- gammaLow[match(LowStates, gammaLow$document), ]
HighStates <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi")
gammaHighNew <- gammaHigh[match(HighStates, gammaHigh$document), ]
gamma_LowHigh <- rbind(gammaHighNew, gammaLowNew)
gamma_LowHigh <- rename(gamma_LowHigh, "gamma_topic_15" = "15")

Vec <- c("nevada", "tennessee", "south-carolina", "new-mexico", "arkansas", "louisiana", "missouri", "alaska", "alabama", "mississippi", "south-dakota", "hawaii", "rhode-island", "north-dakota", "massachusetts", "new-york", "minnesota", "iowa", "connecticut", "oregon")

Top15Prev <- ggplot(data = gamma_LowHigh, mapping = aes(x = factor(document, levels = unique(Vec)), y = gamma_topic_15)) + 
  geom_point(color = "darkred", size = 3) +
  facet_wrap(vars(Hom_Rates), scales = "free_x") +
  labs(title = "Prevalence of Topic 15", 
       subtitle = "for States with Highest and Lowest Female Homicide Rates", x = "State", y = "Prevalence of Topic 15 (gamma)") +
  theme_light(base_size = 10) +
  theme(axis.text.x = element_text(angle = 90))
Top15Prev

Results

The Structural Topic Model allowed for the creation of topics that identify the prevalence of key points of interest in the corpus. Topic 1 represents more passive or fully omitted firearm policy, including 3 key tokens “may_petition_for”, “does_not_prohibit”, and “does_not_require_relinquishment”. This means that documents with a higher prevalence of topic 1 would have less restrictive or firm/directive measures in their firearm legislation. Returning to the research question - “Is there a relationship between the restrictiveness of a state’s firearm regulations/laws for individuals with domestic violence related records and homicide rates perpetrated using a firearm?” - topic 1 does well to identify less restrictive states. Therefore, I proceeded with plotting topic 1 prevalence against the covariate of female homicide rates. As indicated by a positive slope in Plot 1, increasing female homicide rates correlated to increased Topic 1 prevalence. This suggests that states that fail to explicitly prohibit firearm possession for domestic violence offenders, fail to require the relinquishment of weapons, and/or use passive language like “may” instead of, for example, “shall”, tend to experience higher rates of female homicide by firearm. Plot 2 further explores this relationship by allowing the visualization of Topic 1 prevalence in the states with the 10 highest female firearm homicide rates and the 10 lowest rates. The plot shows a marked difference in topic prevalence between the two categories, with almost no prevalence of Topic 1 for the states with the 10 lowest rates. Similarly, Plot 3 displays Topic 1 prevalence in relation to confirmed intimate partner firearm homicide rates for the states that this data is available from. The graph corroborates the findings from Plot 2, which is that states with less restrictive legislative language and provisions tends to have higher intimate partner homicides perpetrated using a firearm. It is important to note that causation is not established, particularly since I did not measure statistical significance nor incorporate control variables. However, the finding establishes significant opportunity for future research exploring the relationship, which will be discussed further in the ‘Takeaways’ section.

Topics 4, 6, and 15, which are the other three topics I decided to explore did not present as strong of an effect (in terms of slope) on homicide rates. Topic 4, which discusses eligibility for restraining orders (in terms of who may file one), showed no positive or negative correlation to homicide rates nor had a greater prevalence in states with higher or lower homicide rates. Topic 6, which is weapon relinquishment, does not appear to be more or less prevalent in states based on homicide rates, but Plot 1 shows a very slight negative slope. If the negative slope was more substantial, it would suggest that states specifying firearm relinquishment measures also experience lower female homicide rates. Lastly, Topic 15, which covers the nature of the relationship between the victim and perpetrator has a slight negative slope as well. This indicates that states with lower female homicide rates have a higher proportion of language in legislation that specifies the relationship between the victim and abuser.

Conclusion and Takeaways

As research continues to reveal the prevalence of firearm usage in intimate partner violence and homicide, it is essential that states work towards establishing comprehensive legislation aimed at reducing high-risk individuals from possessing firearms. Researchers are now tasked with continuing to identify circumstances that are create high-risk of firearm violence, evaluating legislative efficacy, and developing new ways to increase safety for victims of intimate partner violence. As some states continue to implement measures to disarm abusers, others lag behind. Research like this can leverage the differences in legislation across states to evaluate the efficacy of these measures against the absence of measures. As reporting and data collection on intimate partner violence and homicide becomes more prominent and systematic, this improves researchers’ abilities to explore legislation and preventative measures.

For future research, the incorporation of control variables would be essential to producing statistically sound conclusions. Some factors present considerable difficulties to control for in cross-state analysis due to things like jurisdiction differences in resources and implementation. Additionally, it would be useful to analyze the firearm homicide rates in proportion to overall homicide rates by state, to measure how prevalent the use of firearms is in each state. This would allow for a better evaluation of the specificity of firearm legislation, rather than being potentially influenced by other crime factors. Another important iteration to explore is the trend in firearm homicide rates in particular states that have applied greater regulatory legislation. Ideally, the methodology applied to the corpus in this study would be used to evaluate raw firearm legislation by state (with much less pre-processing necessary). Due to each state consolidating and sharing legislation in different ways, it was not feasible for this project to collect that data, but perhaps could be a bridge to cross for my capstone.

Sources and Resources Referenced

• Data Sources:
  • https://wonder.cdc.gov/ucd-icd10.html
  • https://www.cdc.gov/injury/wisqars/nvdrs/
  • https://www.disarmdv.org/#
• Background Research:
  • https://giffords.org/lawcenter/gun-laws/
  • https://everytownresearch.org/rankings/law/prohibition-for-convicted-domestic-abusers/
  • https://www.justice.gov/archives/jm/criminal-resource-manual-1117-restrictions-possession-firearms-individuals-convicted
  • https://www.statefirearmlaws.org/national-data
  • https://nnedv.org/
  • https://www.congress.gov/bill/117th-congress/house-bill/2119
• Topic Modeling and R References:
  • https://content-analysis-with-r.com/6-topic_models.html
  • https://github.com/dondealban/learning-stm
  • https://bookdown.org/valerie_hase/TextasData_HS2021/tutorial-13-topic-modeling.html
  • https://www.tidytextmining.com/topicmodeling.html
  • https://warin.ca/shiny/stm/#section-references
  • https://quanteda.io/articles/quickstart.html
  • https://cran.r-project.org/web/packages/stm/vignettes/stmVignette.pdf