The Relationship Between Indigenous People,Crime,and Extractive Industries in the United States: A Scoping Review

Extractive Industries and Crime

Prior research shows that extractive industries are associated with increases in crime rates in communities compared to the rates prior to the onset of extractive industries or crime rates in similar communities without extractive industries (see systematic review by Stretesky & Grimmer, 2020). Outside of the United States, there has been much more attention to this issue: extractive industries have been associated with sex trafficking in South American countries (U.S. Department of State, 2017), extortion and money laundering, child labor, and labor trafficking in Congo (Zabyelina, 2023), violent crime in the United Kingdom (Stretesky et al., 2018), organized crime and drug trafficking in countries in Latin America (e.g., United Nations Office on Drugs and Crime, 2025), and various forms of transnational and corporate crimes (Maksimentsev & Maksimentsev, 2020), to name a few. Research is much more limited on the topic within the United States, however, and regarding the relationship between extractive industries and crimes against Native Americans specifically, empirical studies on the topic are nearly nonexistent.

The general empirical literature has focused much on the oil and gas industries; Stretesky and Grimmer (2020) reviewed 25 quantitative studies on the relationship between shale gas production and crime rates. Reviewed studies found significant relationships (i.e., p < .05) between shale gas production and rates of total crime (in 4 of 8 studies), violent crime (in 9 of 17 studies), property crime (in 7 of 15 studies), social disorganization crimes (i.e., classified as drug offenses, disorderly conduct, simple assault, drunkenness; in 4 of 6 studies), and violence against women crimes (i.e., rape, domestic abuse, dating violence, stalking, sexual assault; in 7 of 10 studies). However, forestry, gold and other mineral mining, and water and fishing industries have also been of interest. It is suggested that deforestation, for example, can house illegal activities where deforestation takes place and is also associated with activities in illegal markets (Carpio-Dominguez, 2024), while other forms of extractive industries can disrupt local and global financial markets.

The mechanisms underpinning the relationship between the extractive industry are varied, with causality sometimes unclear. For instance, James and Smith (2017) note there is a lack of consensus regarding whether extractive industry “booms attract or produce criminal activity” (p. 126). For example, research has posited that increases in crime rates could be caused by the attraction of large numbers of laborers from groups who are disproportionately likely to commit crimes, such as young men, or more specifically, men who are restricted in the labor market, such as sex offenders (O’Connor, 2014, 2017). Others have argued that changes in crime rates are produced by increases in income inequality and social disorganization (e.g., increases in transient residents/breakdown in community ties) associated with extractive industries or the failure to increase law enforcement resources commensurate with population increases (Stretesky et al., 2018). Similarly, the influx of both people and income creates opportunities for more crimes to occur—bringing both potential perpetrators and victims into an area, alongside increases in cash flow, can create more demand for substance use, alcohol, and other forms of crime (e.g., gambling, theft, disorderly conduct, and so forth, e.g., Carpio-Dominguez, 2024). However, few studies have attempted to test these relationships outside of correlational or associational analyses. One study from James and Smith (2017) examining crime rates among U.S. counties with and without shale gas or tight oil production from 2000 to 2013 found a positive relationship between production and rates of aggravated and simple assault, larceny, grand theft auto, burglary, and forcible rape. However, they found no evidence that counties with oil production experienced disproportionate increases in the percentage of the population who were male or decreases in the number of law enforcement officers per capita. They did, however, find some evidence of a correlation between oil production and the percentage of a county’s population who were registered as a sex offender in some oil-producing counties and a disproportionate rise in income inequality among oil-producing counties compared to control counties. Still, this research relies on correlational relationships and is unable to link perpetrator characteristics (e.g., resident or nonresident) to crimes.

Crime and Violence Against AI/AN People and Responses in Indian Country and Alaska Native Villages

Prior research indicates that AI/AN persons experience crime victimization at higher rates than non-Native people in general (Catalano, 2007; Rosay, 2016), and AI/AN women report experiencing violence in their lifetime, including poly-victimization and severe violence (e.g., violence resulting in injuries, including a weapon) (Bachman et al., 2008; Rosay, 2016) at higher rates than women of other racial/ethnic groups. Thus, Native Americans experience high rates of victimization and serious violence; these issues may become even more potent when considering the impact of extractive industries in these communities.

Tribal governments have limited legal avenues to apprehend, prosecute, and punish people when they perpetrate crimes in Indian country (for a discussion, see Slaff, 2023). To begin, the Major Crimes Act of 1885 (8 U.S.C. § 1153) restricted tribes’ jurisdiction of select crimes (i.e., murder, manslaughter, kidnapping, maiming, sexual abuse, incest, felony assault, felony child abuse and neglect, arson, burglary, larceny, and robbery) committed in Indian country to the federal government regardless of whether the victim is AI/AN. In addition, Public Law 280 (1953) transferred jurisdiction of crimes committed on Indian lands in six states: Alaska, California, Minnesota, Nebraska, Oregon, and Wisconsin, from the federal government to those states. Furthermore, since the Supreme Court’s 2022 decision in Oklahoma v. Castro-Huerta, 597 U.S. 629, state governments (not tribes) now have concurrent jurisdiction with the federal government to prosecute crimes committed by non-Natives against Natives in Indian country. The Indian Civil Rights Act (ICRA, 1968) also placed limitations on tribal jurisdiction regarding sentencing convicted offenders. Under ICRA and its 1986 amendment, a tribal court’s sentencing authority is limited to 1 year in jail or fines up to $5,000, or both. Indigenous legal scholar Deer (2015) argues that these federal laws make violence against Native persons less likely to be punished or prosecuted, and thus, easier to perpetrate and less likely to be reported.

More recently, the federal government has made attempts to address violence in Indian country and Alaska Native Villages through new federal legislation. The Tribal Law and Order Act (2010) (TLOA) enhanced tribes’ sentencing authority over AI/AN offenders and extended access to national criminal justice databases to tribal law enforcement. In addition, the 2013 reauthorization of the Violence Against Women Act (VAWA) established Special Domestic Violence Criminal Jurisdiction (“SDVCJ”) to tribes for domestic violence, dating violence, and violations of protection orders by both Native and non-Native offenders. The recent reauthorization of VAWA (2022; Public Law No. 117-103) expanded the SDVCJ program to include additional crimes and was thus renamed the Special Tribal Criminal Jurisdiction (STCJ) program. The STCJ program expands covered crimes (i.e., assault of tribal justice personnel, domestic and dating violence, child violence, obstruction of justice, sexual violence, sex trafficking, stalking, and violations of protective orders), as well as jurisdiction over all defendants (both Indian and non-Indian) committed in Indian country and Alaska Native Villages. However, the STCJ does not address crimes perpetrated by non-Native acquaintances or strangers against Native victims in Indian country and Alaska Native Villages (see discussion in Gilbert et al., 2021), which may be particularly relevant to the relationship between crime in Indian country and Alaska Native Villages and extractive industries, as extractive industries often result in an influx of new workers to an area. Furthermore, as of July 2025, the Department of Justice was aware of only 38 federally recognized tribes that had implemented STCJ (See Supplemental File A).

Land Regulation and Extractive Industries in Indian Country and Alaska Native Villages

The U.S. federal government has had a regulatory relationship with tribal lands and land resources since treaty-making began in the late 1700s. In these treaties, tribes ceded the rights, title, and interests in their lands to the United States while retaining or reserving certain rights. The Supreme Court has upheld these treaties, and the rights granted to tribes, such as rights to water (e.g., United States v. Winans, 198 U.S. 371, [1905; see also, Winters v. United States, 207 U.S. 564 [1908]) or minerals (United States v. Shoshone Tribe of Indians, 304 U.S. 111 [1938]).

Beginning in 1850, a series of treaties, statutes, and executive orders mandated that tribes move onto reservation lands (i.e., land reserved for the tribe) (U.S. Department of the Interior Indian Affairs, n.d.). For some tribes, reservations were distant from their homeland, while for others, they comprised a small area of their original territory (U.S. Department of the Interior, Indian Affairs, n.d.). Furthermore, the General Allotment Act of 1887 divided reservation land into individual allotments for each member of a tribe, while the remaining reservation land was designated as “surplus” by the federal government and opened to non-Native homesteaders (U.S. Department of the Interior Indian Affairs, n.d.). As a result, tribal-owned land was dramatically reduced from 138 million acres in 1887 to 48 million acres in 1934 (U.S. Department of the Interior Natural Resources and Revenue Data, n.d.). Furthermore, as tribal recipients of land allotments have lost their land to unpaid taxes, sold their land, or whose decedents now own fractions of original allotments (i.e., fractionation), many reservations now exist as a “checkerboard of land ownership” with both Native and non-Native landowners as well as parcels of land with many co-owners (U.S. Department of the Interior Natural Resources and Revenue Data, n.d.) complicates jurisdictional challenges (e.g., split authority, regulatory gaps) and accountability (e.g., limited control over contiguous land areas, making governance and prosecution difficult).

In Alaska specifically, the Alaska Native Claims Settlement Act (ANCSA) of 1971 extinguished aboriginal land title and created 12 for-profit Alaska Native regional corporations and more than 200 private Alaska Native village corporations—all owned by enrolled Alaska Native shareholders (ANCSA Regional Association, n.d.; U.S. Department of the Interior Natural Resources Bureau of Land Management, n.d.). Through ANCSA, more than 44 million acres of land were transferred by the federal government to more than 74,000 Alaska Native regional and village corporation shareholders (ANCSA Regional Association, n.d.). Over time, many Alaska Native corporations have engaged in land transfers with each other as well as with state and federal agencies, resulting in a mix of legal rights to surface (i.e., above the ground) and subsurface (i.e., below the ground) resources across Alaska Native regional and village corporations as well as non-Native government agencies (ANCSA Regional Association, n.d.).

A series of federal laws allows the use of tribal lands for extractive industries for mining, forestry, or other purposes, with input by the tribe(s) affected. To demonstrate, the Indian Mineral Leasing Act of 1938 (IMLA) allows tribes to lease their lands out for extraction, with tribal consent. However, they are limited in the negotiation terms they can set (e.g., rates, lease cancellation), and the U.S. Secretary of the Interior must approve the terms. In addition, the Indian Mineral Development Act of 1982 (IMDA) allows tribes and tribal individuals to negotiate their own “mineral development agreements” (MDAs) with extraction companies, but like the IMLA, it requires approval by the U.S. Secretary of the Interior.

Furthermore, tribes’ sacred religious sites (and sacred objects) are protected by the American Indian Religious Freedom Act (AIRFA) of 1978 and may limit extractive industries’ access to certain locations. Similarly, the National Historic Preservation Act (NHPA) of 1966 established a partnership between the federal government and state, tribal, and local governments for preservation activities and created the Advisory Council on Historic Preservation (ACHP) to address historic preservation issues. This is relevant to tribal lands because it seeks to preserve lands of historic relevance—many of which are found on or near reservations. Section 106 of the NHPA requires that federal agencies must consider the potential impact of the extraction on historical and protected places, and consult with the affected communities (in this case, tribes) about the project (these projects are then reviewed by the ACHP). However, the NHPA only encourages preservation; it does not mandate it. Thus, while the U.S. federal government serves as a trusted caretaker of Native lands, there are some caveats to this protection: while tribes should have ownership or access to sacred lands and objects, and those areas should be preserved and protected, the protections to tribal land and resources are dependent upon justifications, consultations, and approvals from multiple entities, both tribal and nontribal.

Taken together, natural resources often found on or near tribal reservations, combined with the regulatory gaps and split authority in legal protections of these lands, increase the likelihood of extractive industries in Indian country and Alaska Native Villages. Even when AI/AN people and tribes potentially benefit from economic and development opportunities, extractive industries may also have negative impacts, such as increases in crime and violence in tribal communities that are exacerbated because of jurisdictional complexities. If a crime occurs in Indian country or an Alaska Native Village, the jurisdictional authority for apprehending and punishing the perpetrator is dependent on the crime that was committed, whether the offender and/or the victim are Native, and whether the crime was committed exclusively in Indian country or an Alaska Native Village (see Castillo, 2015; Mikkanen, 2020). To the extent that these industries import potential perpetrators of crime (whether Native or non-Native, tribal member or not), tribes’ abilities to protect their members from crime remain limited.

Current Study

In the current scoping review, we examine the existing research on the relationship between extractive industries and crime and violence experienced by AI/AN people. A scoping review is a “systematic approach to map evidence on a topic and identify main concepts, theories, sources, and knowledge gaps” (Tricco et al., 2018, p. 467). Scoping reviews can be useful in an emerging field to determine the “scope” of a body of literature on a topical area and to report the types of evidence available and the ways such evidence has been collected (Munn et al., 2018). In particular, we investigated the following research questions:

Ultimately, we hope to inform researchers, policymakers, and practitioners about the current state of knowledge and methods of knowledge gathering so that they can make informed decisions about future efforts regarding research, policy/legislation, and practices pertaining to extractive industries and tribal communities.

Stages of Scoping Review

Below, we describe details on each of the four stages included in the current scoping review: (a) establishing eligibility criteria; (b) identifying relevant studies; (c) screening relevant studies; and (d) charting and organizing data. Our protocol was preregistered on Open Science Framework at https://doi.org/10.17605/OSF.IO/GF6SX.

Eligibility Criteria

Studies were considered eligible for inclusion if they (a) reported findings from an empirical study (i.e., not literature reviews, theoretical articles, commentaries, or conceptual papers); (b) were written in English; and (c) examined extractive industries and crime and violence experienced by Indigenous peoples in the United States. No studies were excluded based on the year of study. Unpublished theses, dissertations, and reports were also included.

Identification of Relevant Studies

MeSH and other terms combined with Boolean Operators (OR and AND) were used to create the following search equation: (“homicide” OR “murder” OR “death investigations” OR “intimate partner violence” OR “domestic violence” OR “sexual violence” OR “missing and murdered” OR “gender violence” OR “family violence” OR crime OR violence OR trafficking OR rape OR assault OR “sex crimes”) AND (“resource booms” OR “man camps” OR fracking OR boomtown* OR “oil development” OR “energy development” OR “extract* industries” OR “natural resource*” OR mining OR “shale development” OR “unconventional oil and gas” OR “oil and gas Industry”) AND (American Indians or Alaska Natives OR Native American OR Indigenous). The search equation was developed by first examining search terms from a previous, related systematic review from the National Criminal Justice Reference Service (NCJRS) Library and then revised for this project with the assistance of a reference librarian at the University of Nebraska Omaha. A draft protocol was submitted to a scientist at the National Institute of Justice for feedback, and their feedback was integrated into this finalized protocol.

The search equation was run for title, abstract, and keywords in four electronic databases: PubMed, PsycINFO, SCOPUS, and Web of Science, and in abstract and summary text in ProQuest Dissertations and Theses and ProQuest Social Science Premium Collection (which includes NCJRS Abstracts). We also search Google Scholar using a combination of “Native and violence and extractive industry” and “Indigenous and violence and extractive industry.” The first 100 records sorted by relevance were reviewed from Google Scholar. In addition, we attempted to contact scholars from the Native American Research Centers for Health (NARCH), the 12 Tribal Epidemiology Centers, the Alaska Native Tribal Health Consortium, the Society of Indian Psychologists, and the Native Research Network (NRN) to assist with the identification of unpublished reports; we were ultimately successful in identifying email addresses for the NARCH, nine of the Tribal Epidemiology Centers, the Alaska Native Tribal Health Consortium, and the Society of Indian Psychologists. We received responses from the Society of Indian Psychologists and one of the Tribal Epidemiology Centers. We also reviewed the reference lists of included studies for additional records.

Selection of Studies

All articles were reviewed, and after removing duplicates, the authors independently screened articles based on the title and abstract (Stage 1 Review). Disagreements in screening were resolved using a third screener, a doctoral-level research assistant. Studies “screened in” at Stage One were read in full by one of the authors and the doctoral-level researcher (Stage 2 Review). Again, disagreements were resolved using a third screener (i.e., the other author). Inclusion/exclusion was tracked at each stage using the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow chart (see Figure 1). Rayyan software was used to manage and organize the review process.

OPEN IN VIEWER

Figure 1.

PRISMA 2020 flow diagram for scoping review on extractive industries and crime and victimization experienced by indigenous populations.

Stage 1: Title and Abstract Review

Using the above search terms, 2,156 articles were identified. After removing 213 duplicates manually and using the duplicate detection function in Rayyan (with the recommended threshold of 96% similarity), 1,943 articles were included in the Stage 1 review (title and abstract review). In the Stage 1 review, 1,917 articles were excluded because they did not meet the inclusion criteria, leaving 26 articles for the Stage 2 review (i.e., full-text review).

Stage 2: Full-Text Review

In Stage 2, the 26 remaining articles were read in full; 4 articles were removed because they were not empirical studies; 19 articles were removed because they did not meet the topical inclusion criteria (i.e., they were not focused on extractive industries and crime and violence among AI/AN persons). The remaining three articles were included in this scoping review. In addition, the reference pages for these three articles were reviewed for additional records that met the inclusion criteria; two additional records, both National Institute of Justice-funded studies, were reviewed and included in the scoping review.

Data Charting

Two reviewers (i.e., a trained doctoral student and one of the authors) conducted extraction on all five included studies, with a third reviewer (i.e., the other author) ensuring consensus was reached. The data were charted using a structured data extraction spreadsheet in Microsoft Excel to systematically capture the following information: (a) author(s), (b) year, (c) study location, (d) sample characteristics (e.g., tribal affiliation/s, gender, age), (e) extractive industry/ies of focus, (f) crime type/s, (g) sources/type of data used, (h) methods used, (i) main findings, and (g) limitations.

Results

Five records were included in this scoping review: two peer-reviewed journal articles from International and Environmental Studies (i.e., Chase & Johnson, 2024), and Forensic Science (Joseph, 2021), respectively; a master’s thesis from International Environmental and Development Studies (Rogers, 2017), and two final reports from National Institute of Justice-funded studies (i.e., Jayasundara et al., 2016; Martin et al., 2019). Publication dates ranged from 2016 to 2024. Three of the articles focused on the “Bakken region” (i.e., land that contains the Bakken oil shale formation), but each article defined the area slightly differently. Chase and Johnson (2024) identified the Bakken region as “~200,000 square miles in North Dakota and Montana, USA, and Saskatchewan and Manitoba, Canada,” while Martin et al. (2019) defined it as “counties in Montana and North Dakota that contain the Bakken shale formation,” and Jayasundara et al. (2016) specified “12 Montana and 21 North Dakota counties.” The remaining two articles focused on the Standing Rock Sioux Reservation (i.e., where the Dakota Access Pipeline is located; Rogers 2017), and U.S. states, cities, and counties where murdered or unidentified remains/missing Indigenous women/girls were identified (Joseph, 2021).

The first research question asked about the data and methods used to investigate the relationship between extractive industries and crime and violence experienced by Indigenous peoples in the United States. To begin, none of the included studies provided robust information about the characteristics of their samples. For example, Jayasundara et al. (2016) categorized their interview and focus group participants (e.g., “general human service providers,” “healthcare providers,” “victims”) and reported the number of focus group and interview participants in each category, respectively, and the number of tribal members in each category, but no other information was provided. Chase and Johnson (2024) reported that their interviewees comprised, “tribal members of the Northern Great Plains who were spiritual and political leaders, residents, oil industry workers, and employees or retirees from different tribal government departments, stay-at home parents, members who grew up on the reservation but had recently relocated, tribal beneficiaries of oil royalties, and members of other native tribes that were in the Bakken region for employment” (p. 34); however, they did not report the number of participants in each group or any other information. Rogers (2017) reported that of their six interviewees, two were Indigenous and four were non-Native, three were in their 20s, one in their 30s, and two in their 40s, and that four were males (the gender of the remaining two participants was not reported). Joseph (2021) did not provide any information on the characteristics of their sample of unidentified remains and missing persons other than the inclusion criteria (i.e., that they were Indigenous and female), and they reported these inclusion criteria, as well as age, for their sample of unidentified remains. Finally, neither Jayasundara et al. (2016) nor Martin et al. (2019) provided sample characteristics for any of the secondary data they utilized (e.g., from UCR, NIBRS, state-level data); however, Jayasundra and colleagues conducted all analyses at the region level.

Regarding types of extractive industries examined, three of the studies focused on shale oil production, one on hydraulic fracking, and one on the oil industry (i.e., the Dakota Access Pipeline; Rogers, 2017). All studies addressed violent crime victimization with three papers focused specifically on violence against Indigenous women (i.e., Chase & Johnson, 2024; Jayasundara et al., 2016; Joseph, 2021): one paper focused on violent crimes included in NIBRS (i.e., homicide/non-negligent manslaughter, rape/sexual assault, unlawful sexual a contact, robbery, aggravated assault, simple assault, other violent offenses), and one paper focused on “direct, cultural, and structural violence” where direct violence comprised “killing, harming the body, or restricting movements or the mental functions of a person” (Rogers, 2017, p. 4) (see Table 1). It should also be noted that while Joseph (2021) included data on murdered Indigenous women in their study, the bulk of their analyses focused on missing persons reports and unidentified remains, as going missing is not a crime and does not necessarily mean that violence or victimization has occurred (see Chakraborty, 2019; Richards et al., 2021).

OPEN IN VIEWER

Table 1.

Descriptive Information for the Articles Included in the Review.

Study (authors, year)	Study location	Sample	Extractive industry/ies	Crime types
Chase and Johnson (2024)	Bakken region is defined here ~200,000 square miles in North Dakota and Montana, USA; Saskatchewan and Manitoba, Canada	Study participants (n = 25) included tribal members of the Northern Great Plains who were spiritual and political leaders, residents, oil industry workers, and employees or retirees from different tribal government departments, stay-at-home parents, members who grew up on the reservation but had recently relocated, tribal beneficiaries of oil royalties, and members of other native tribes that were in the Bakken region for employment	Shale oil production	Sexual assault and violence against Indigenous women
Rogers (2017)	Standing Rock Sioux Reservation, South Dakota, and North Dakota during the DAPL conflict	Study participants (n = 6) were all adult men; two were Indigenous, and four were non-native; all had personally been to Standing Rock during the DAPL conflict	Oil industry/DAPL	Direct, cultural, and structural violence are addressed. Direct violence can manifest in psychological and physical ways involving killing, harming the body, or restricting movements or the mental functions of a person and the perpetrator is known; structural violence refers to the institutional structures in place, which can cause loss of dignity and opportunity and barriers to gaining a livelihood; and cultural violence is directed toward varying cultural aspects, including religion as a justification
Joseph (2021)	States, counties, and cities across the United States have reported cases of murdered and/or missing Indigenous women	American Indian/Alaska Native women/girls who were listed as missing or murdered in the Urban Indian Health Institute report or as unidentified remains (n = 24) or missing persons (n = 153) in NamUS on September 9, 2019	Hydraulic fracking	MMIW
Martin et al. (2019)	Counties in Montana and North Dakota that contain the Bakken shale formation (i.e., Bakken counties/region), as well as counties in Montana, North Dakota, and South Dakota that do not contain the Bakken shale formation (i.e., Bakken counties/region)	NIBRS violent crime incidents for Montana, North Dakota, and South Dakota from 2006 to 2012	Shale oil production	Violent crimes reported to NIBRS: homicide/non-negligent manslaughter, rape/sexual assault, unlawful sexual contact, robbery, aggravated assault, simple assault, other violent offenses
Jayasundara et al. (2016)	Bakken Region defined here as 12 MT and 21 ND counties, including among the Mandan, Hidatsa, and Arikara Nation, Fort Peck Assiniboine and Sioux Tribes, and Turtle Mountain Band of Chippewa	Individuals in sampled MT and ND counties from 2004 to 2015 (i.e., for victimization data); community members and relevant professionals (e.g., criminal justice personnel, oil industry workers); 27.6% participants were interviewed in tribal communities, 21.1% of participants were affiliated with a tribe in the region (for interviews and focus groups)	Shale oil production	DV, DatV, SA, and ST

Note. DAPL = Dakota Access Pipeline; MMIW = Missing and murdered Indigenous women; NIBRS = National Incident Based Reporting System; MT = Montana; ND = North Dakota; DV = domestic violence; DatV = dating violence; SA = sexual assault; ST = stalking.

Two of the studies utilized only qualitative data (Chase & Johnson, 2024; Rogers, 2017), two studies utilized only quantitative data (Joseph, 2021; Martin et al., 2019), and one study used both qualitative and quantitative data (Jayasundara et al., 2016). Qualitative data stemmed from interviews with tribal and non-tribal community members and individuals associated with the oil industry (n = 25) (Chase & Johnson, 2024) and individuals who had been to the Standing Rock Sioux Reservation during the Dakota Access Pipeline conflict (n = 6) (Rogers, 2017), as well as interviews (n = 119) and focus groups (n = 66) with community members and professionals from the oil industry (i.e., victim, health, human services, and the criminal justice system) (Jayasundara et al., 2016). Two studies specifically noted that the qualitative data (i.e., interview transcripts) were analyzed using grounded theory (Chase & Johnson, 2024; Rogers, 2017). The third study reported using NVivo software to analyze the qualitative data by “reading and ‘coding’ the interviews by identifying reoccurring themes and noting sources of variation between different contributors” (Jayasundara et al., 2016, p. 5).

The quantitative data were sourced from multiple secondary datasets, including the National Missing and Unidentified Persons System (NamUs), National Incident Based Reporting System (NIBRS), Uniform Crime Report (UCR), and a previous report on missing and murdered Indigenous women/girls in 71 U.S. Cities (Urban Indian Health Institute, 2018), as well as individual state- and community-based agency data sources. Methods to analyze quantitative data included descriptive statistics (i.e., frequencies, proportions, and percentages) and bivariate analyses, including t-tests and chi-square tests. Two studies employed quasi-experimental designs that compared counties’ pre-extractive industry and post-extractive industry and employed a treatment group (i.e., counties with extractive industries) and a control group (i.e., adjacent counties without extractive industries) (i.e., Jayasundara et al., 2016; Martin et al., 2019).

Our second research question sought to understand what is known regarding the relationship between extractive industries and crime and violence experienced by AI/AN people in the United States (see Table 2). As noted above, three of the studies included here focus on the Bakken region. The first study compared UCR and NIBRS data from 2002 to 2014 to examine whether there were statistically significant differences in the average number of victimizations reported prior to the oil boom (2004–2007) compared to after the oil boom (2009–2014) for 33 counties in the Bakken region, including the Mandan, Hidatsa, and Arikara Nation, Fort Peck Assiniboine and Sioux Tribes, and Turtle Mountain Band of Chippewa (Jayasundara et al., 2016). Findings showed significant increases in the number of NIBRS-reported domestic violence (DV) and dating violence incidents, and the number of reports for a combined overall DV, dating violence, sexual assault, and stalking incident composite measure from the pre-oil boom period to the post-oil boom period. Furthermore, the authors found no significant differences in the number of NIBRS-reported sexual assault, stalking, or intimidation incidents pre- versus post-oil boom. When the authors compared data for departments within the Bakken region (n = 20) and outside the Bakken region (n = 90), they found that the annual rates of NIBRS-reported DV, dating violence, and the overall composite measure in the Bakken region increased from 2005 to 2014, while in the non-Bakken region, annual rates of NIBRS-reported dating violence increased while annual rates of DV and the overall composite measure decreased. Finally, Jayasundara and colleagues examined the number of sexual assault victims served by the programs in the North Dakota Council on Abused Women’s Services and found a “modest increase” in average annual numbers of victims served in the Bakken region from 2002–2007 compared to 2009–2014, but not for the non-Bakken region. No analyses were conducted comparing tribal and non-tribal areas or American Indian/Alaska Native victims compared to victims who were not American Indian/Alaska Native.

OPEN IN VIEWER

Table 2.

Data, Methods, Key Findings, and Limitations for the Records Included in the Review.

Data sources	Methods	Key findings	Limitations
Conversational interviews (consistent with Indigenous methodologies); participatory observations at community meetings, at search efforts for missing women, local restaurants, shops, gas stations, and the central Wal-Mart.	Transcribed interviews and observation notes were coded following principles of grounded theory, where emergent themes are first identified and coded, and theory is constructed out of the most applicable categories. Data were analyzed to address two research questions (1) What are the necropolitical settler tactics of oil extraction that make Indigenous women particularly vulnerable to violence and attrition? and (2) What are their economic, gendered, and jurisdictional spatialities in the Bakken region?	The authors note that factors contributing to violence in the region include economic inequality stemming from structural poverty, uneven oil wealth distribution, and intertribal wealth disparities, which increase financial dependence on extractive industries and consequently, vulnerability to violence. Additionally, jurisdictional challenges arise from overlapping tribal and federal laws that limit tribal authority to prosecute non-Native offenders. The authors further suggest that “man-camps”—pop-up communities that house mostly male, transient oil workers–lead to sexual exploitation of women in the community. The authors conclude that to address these issues, solutions must include the restoration of tribal sovereignty, the promotion of Indigenous-led initiatives, national tracking systems for cases of missing and/or murdered Indigenous women, cultural healing, and reduced reliance on federal institutions that perpetuate systemic violence.	Exploratory research. Small, convenience sample. Minimal information on participants.
Semi-structured interviews; secondary data from social media and websites connected to the Standing Rock Sioux Tribe/other Sioux tribes, websites connected to Energy Transfer Partners and the Dakota Access Pipeline, documents from the Army Corps of Engineers and the United Nations, video footage from the United Nations Permanent Forum on Indigenous Issues and various Indigenous and mainstream media outlets, and a full-length film on the conflict at Standing Rock.	Semi-structured interviews with key informants and qualitative analysis of secondary data were used to examine a series of research questions, including one relevant to this review (1) How is direct violence manifested in the conflict at Standing Rock? This study employs a qualitative approach in grounded theory and adopts an emotionalist perspective, emphasizing subjectivity, inner human experiences, and access to participants’ lived realities.	The author purports that the DAPL highlights the issues of extractive politics, noting that the dispossession at Standing Rock includes the loss of land, health, and cultural identity. Interviewees noted that the DAPL representatives disregarded the concerns of the Standing Rock Sioux Tribe, undermining their right to oppose harmful environmental developments impacting sovereign tribal nations. Findings on direct violence against members of the Standing Rock Sioux Tribe and protestors included the use of rubber bullets, dogs, water cannons; structural violence manifested through systemic inequities, including the economic benefits from DAPL bypassing the Standing Rock Sioux Tribe and violations of their treaties; and cultural violence included the failure to uphold Indigenous rights and self-determination (e.g., lack of consultation with the Tribe).	Exploratory research. Small, convenience sample. Minimal information on participants and how they were recruited into the study.
Number of cases of murdered Indigenous women from Urban Indian Health Institute Report; number of unidentified remains/missing persons, sex, age, location of unidentified remains/missing persons from NamUS; population data for sampled cities/counties from U.S. Census; active fracking sites from Earth Justice, Oil and Gas Threat Map, and Drilling Maps.	The number of cases of missing or murdered Indigenous women across cities/states from the Urban Indian Health Institute Report was identified. The number of cases of missing/unidentified remains of Indigenous women in NamUS was conducted. Missing persons data by location were examined to determine whether missing AI/AN women cases were concentrated in specific locations (i.e., locations with three or more were considered “hotspots”). Census data for locations with hotspots from NamUS were analyzed to assess whether the proportion of missing Indigenous women/girls was higher than expected based on their representation in the population. Chi-square tests and z-tests were used to compare actual and expected rates of missing Indigenous women. Fracking/drilling maps were overlaid with “hotspots” to examine the geographic relationship between those areas.	According to NamUS, the states with the highest numbers of unidentified remains/missing Indigenous women/girls were AK (20), AZ (11), CA (13), NM (12), OK (24), and WA (20); and top five cities/counties were Albuquerque (NM), Anchorage (AK), Gallup/McKinley County (NM), Seattle/King County (WA), and Tacoma/Pierce County (WA). AK, NM, NC, and WA had significantly higher rates of unidentified remains/missing Indigenous women/girls than expected, given their representation in the population; seven cities/counties had significantly higher rates than expected. Tuscon and Pheonix (AZ), Salt Lake City (UT), Billings, (MT), Minneapolis/St. Paul (MN), Denver (CO), Chicago (IL), and Oklahoma City (OK) were additional MMIW hotspots added from the UIHI report; 23 locations were identified as hotspots. Six hotspots were within 25 miles of an active fracking/drilling site, six within 50–156 miles, and the remaining 8 were within 157 miles.	Exploratory research. Missingness is not necessarily associated with violence or victimization. No rational for using “3” as the threshold for “hotpots” or scale of distance measure (i.e., 25 miles or less, 50–156 miles).
NIBRS data years 2006–2012; 71% of agencies reported at least 6 months of data to NIBRS each year in 2006 and increased to 86% in 2012; tribal agencies do not report to NIBRS.	Region-level crime rates in 2006–2008 (pre-oil production) were compared to crime rates in 2009–2012 (during oil production); rates were also disaggregated by victim/offender relationship, race, age, victim sex, and by victim sex/offender sex dyads. Percent change in rates was also calculated. Imputation was used to address missing data. Research questions included (1) whether violent crimes reported to law enforcement in the Bakken oil region increased in frequency and severity after 2008 and (2) if the rise in violent crimes reported by law enforcement in the Bakken region was part of broader geographic changes in violent crime that extended to non-oil-producing counties outside the Bakken region.	From 2006 to 2012, the rate of violent victimizations in the Bakken region increased 23% from 105.3 per 10,000 persons in 2006 to 129.9 per 10,000 persons in 2012, compared to a decrease of 8% in the non-Bakken region (from 135.4 to 124 per 10,000). In the Bakken region, the violent victimization rate increased over the study period for every type of violent offense other than rape and sexual assault. The greatest increase was experienced for rates of aggravated assault and unlawful sexual contact (70.1% and 44.8%). Women had a higher victimization rate across the study period, but the percent change in violent victimizations in 2006 compared to 2012 was higher for men/boys (30.8% change) compared to women/girls (18.3% change); men/boy’s and women/girl’s increases in victimization primarily stemmed from male perpetrators (29.8% and 19.4% change) and strangers (50.5% and 29.0% change); however, domestic violence increased for both male and female victims (25.2% and 29.0% change). In the Bakken region, individuals identifying as Black and Native American had higher victimization rates than Whites across the study period; all races saw increased rates of victimization from 2006 to 2012 with Whites and Native Americans experiencing the greatest increases (24.4% and 23.5% change) compared to Blacks (9.8% change). Regarding age, adults 20–24 years old had the highest victimization rates across the study period, but adults 50-64 and 35–49 experienced the greatest increases in victimization (67.7% and 56.7% change).	Most methodologically rigorous study: used a pre-post design and comparison group/region, and rigorous imputation for missing data. However, tribal agencies do not report to NIBRS, so the number of missing crimes in tribal jurisdiction is unknown, likely omitting crimes against AI/AN persons and committed by non-AI/AN persons (given known data on AI/AN victims/perpetrators).
Data on population demographics from the U.S. Census (2000–2014); registered sex offenders from the ND Attorney General (2010–2015) and MT Department of Justice (2008–2014); DV, DatV, SA, ST from Uniform Crime Report summary statistics on rape (2002–2014); National Incident Based Reporting System-victim-level data on DV, DatV, SA, and intimidation/stalking (2002–2014); ND Council on Abused Women’s Services (ND CAWS) statistics on the number of DV and SA victims served by their programs (2001–2014); and Automated Victims’ Information Database (AVID) information maintained by the MT Board of Crime Control (MBCC) on the number of clients with primary victimizations classified as DV or partner/family member assault, SA, and ST served by various funded programs (2010–2015). UCR and NIBRS data on populations served by specific law enforcement agencies were also used for rate computations. Semi-structured interviews (n = 119) and focus groups (n = 66) with community members and professionals.	Annual caseloads for DV, DatV, SA, and ST crime and victimization data were analyzed at the agency or program level; data on population demographics from the Census and on registered sex offenders were analyzed at the county level. Repeated measures t-tests were conducted to determine if there were any statistically significant differences in the average number of victimizations reported or clients served prior to the oil boom compared to after the oil boom was underway. Victimization rates for population changes were also examined. Analyses were conducted for agencies/programs both within and outside of the Bakken region; population change was visualized using hotspot mapping. Interviews and focus groups were recorded and transcribed; transcripts were coded for themes and similarities/differences across participants. Relevant research questions included (1) Have incidents of DV, DatV, SA, and ST increased since the oil boom; (2) Are any observed changes commensurate with population growth; (3) Are any changes unique to oil-impacted communities; (4) Can maps depict the relationship between crime hotspots and changing population demographics; (5) What are the characteristics of victims and offenders and the nature of victims’ experiences; and (6) How has oil development impacted perceptions of DV, DatV, SA, and ST, among victims, criminal justice personnel, direct victim service providers, healthcare providers, general human service providers, tribal community members, oil industry personnel, and community members.	In the Bakken region, there were significant increases in the number of NIBRS-reported DV, DatV, and an overall DV, DatV, SA, ST composite from 2004–2007 to 2009–2014; there was no significant change in the number of NIBRS-reported SA and intimidation. There was a “modest increase” in average annual numbers of SA victims served by ND CAWS across ND’s Bakken region from 2002–2007 to 2009–2014. Law enforcement agency and victim service program “hotspots” for DV, DatV, SA, and ST and DV victims, respectively, were identified in ND hub cities and/or near the epicenter of the oil boom. Furthermore, in the Bakken region, annual rates of NIBRS-reported DV and DatV and an overall DV, DatV, SA, ST composite measure increased from 2005 to 2014. In the non-Bakken region, annual numbers of NIBRS-reported DatV increased while annual rates of DV and rates of an overall DV, DatV, SA, ST composite measure decreased from 2005 to 2014. Hotspot mapping also suggested population growth and an increase in registered sex offenders and men in the Bakken region during the oil boom. Interviews with professional noted changes in their service population with more new victim-clients/offenders; more cases involving alcohol, drug use, and/or mental health; more cases of child abuse/neglect, and growing financial dependency of women on men (with increasing oil wages). Furthermore, existing oil company policies focused on drug use and workplace violence and did not address family violence; fears about employment repercussions (e.g., being fired, losing oil company housing) may have deterred some victims from reporting violence.	Exploratory research. Used a pre-post design and comparison group/region and attempts to triangulate findings across multiple data sources. However, the data elements across sources are not well defined, the operationalization of measures is not explained, and the dates of available data across sources differ. Overall, the methods are too brief to fully assess the rigor of the study.

Note. DAPL = Dakota Access Pipeline; DV = domestic violence; DatV = dating violence; SA = sexual assault; ST = stalking.

Building on Jayasundara et al. (2016), Martin et al. (2019) used NIBRS data to compare crime rates in 2006 to 2008 (pre-oil production) with rates in 2009 to 2012 (during oil production) for the Bakken and non-Bakken region; rates were also disaggregated by victim/offender relationship, race, age, victim sex, and by victim sex/offender sex dyads. Findings showed that the overall rate of violent victimization in the Bakken region increased 23% from 2006 to 2012 compared to a decrease of 8% in the non-Bakken region. Furthermore, in the Bakken region, the violent victimization rate increased over the study period for every type of violent offense other than rape and sexual assault; the greatest increases were for rates of aggravated assault and unlawful sexual contact. In the Bakken region, individuals identifying as Black and Native American had higher victimization rates than Whites across the entire study period; all races saw increased rates of victimization from 2006 to 2012, with Whites and Native Americans experiencing the greatest increases (24.4% and 23.5% change, respectively) compared to Blacks (9.8% change). No other analyses were performed by race, so it is not possible to compare the context of victimization: victim/offender relationship, victim age, victim sex, and victim sex/offender sex dyads for American Indian/Alaska Native victims or examine potential differences for American Indian/Alaska Native victims compared to victims who were not American Indian/Alaska Native.

Most recently, Chase and Johnson’s (2024) article relies on news reports to assert that violence against Indigenous women has increased in the Bakken region because of the oil boom and uses information gleaned from 25 interviews to examine the underlying causes of this increase. They suggest that extractive industries contribute to violence against American Indian/Alaska Native people through increased economic inequality. Specifically, they note that uneven oil wealth distribution and intertribal wealth disparities increases AI/AN peoples’ and tribes’ financial dependence on extractive industries and thus, their vulnerability to violence. The authors further suggest that “man-camps” lead to sexual exploitation of women in the community and that the limitations of tribal authority to prosecute violence against Indigenous women by non-Native perpetrators leave the region in “conditions where culprits, likely connected to the oil industry, can function with impunity” (Chase & Johnson, 2024, p. 37). However, this article does not directly test any of these arguments.

Furthermore, Joseph (2021) attempted to examine the relationship between the “missing and murdered Indigenous women (MMIW) crisis. . .hotspots and fracking locations in the US” (p. 2). The study used the number of homicides of Indigenous women and missing Indigenous women in 71 U.S. cities from a report by the Urban Indian Health Institute, as well as missing persons reports and cases of unidentified remains in NamUS, as measures of MMIW. Across these data sources, 23 locations were identified as MMIW “hotspots” (i.e., locations with three or more MMIW cases). The author notes that six hotspots were within 25 miles of an active fracking/drilling site, six were within 50 to 156 miles, and the remaining 8 were within 157 miles.

Finally, Rogers (2017) used interviews with six individuals who had personally been to the Standing Rock Sioux Reservation during the Dakota Access Pipeline (DAPL) conflict and a range of secondary data sources to examine “direct, cultural, and structural violence” (p. 36) against Indigenous persons perpetrated during the conflict. The author purports that the DAPL conflict highlights issues of extractive politics, noting that the dispossession at Standing Rock includes the loss of land, health, and cultural identity. Interviewees reported direct violence against members of the Standing Rock Sioux Tribe and protestors, including the use of rubber bullets, dogs, and water cannons. Structural violence manifested through systemic inequities, including the economic benefits from DAPL bypassing the Standing Rock Sioux Tribe and violations of their treaties. Finally, cultural violence included the failure to uphold Indigenous rights and self-determination (e.g., lack of consultation with the tribe).