Skip to main content
NCBI home page
Elsevier - PMC COVID-19 Collection logoLink to Elsevier - PMC COVID-19 Collection
. 2023 Mar 6;86:102050. doi: 10.1016/j.jcrimjus.2023.102050

Does routine activity theory still matter during COVID-19 restrictions? The geography of sexual assaults before, during, and after COVID-19 restrictions

Kim M Lersch a,⁎, Timothy C Hart b
PMCID: PMC9986134  PMID: 36911596

Abstract

On March 10, 2020, the Governor of the State of Michigan, USA, declared a state of emergency in response to COVID-19. Within days, schools were closed; in-person dining was restricted; and lockdowns and precautionary stay-at-home orders were issued. These restrictions dramatically impacted the mobility of offenders and victims through space and time. As routine activities were forced to change and crime generators were closed, did hot spots and risky locations for victimization change as well? The purpose of this research is to analyze potential shifts in high-risk areas for sexual assaults before, during, and after COVID-19 restrictions. Using data from the City of Detroit, Michigan, USA, optimized hot spot analysis and Risk Terrain Modeling (RTM) were used to identify critical spatial factors for the occurrence of sexual assaults before, during, and after COVID-19 restrictions. The results suggested that hot spot areas for sexual assaults were more concentrated during the COVID timeframe, compared to the Pre-COVID timeframe. While blight complaints, public transit stops, points of sale for liquor, and the locations of drug arrests were consistent risk factors for sexual assaults before and after COVID restrictions, other factors, such as casinos and demolitions, were only influential in the COVID period.

Keywords: Sexual assault, Rape, Risk terrain modeling, COVID-19, Environmental criminology

1. Introduction

The first case of a new coronavirus, COVID-19, was reported in the United States on January 20, 2020. By May 28th, 2020, there were nearly 1.7 million reported cases and over 100,000 deaths attributed to the virus (Centers for Disease Control and Prevention, 2022, Centers for Disease Control and Prevention, 2023). The pandemic is far from over. As of February 1, 2023, over 102 million cases have been reported with over 1.1 million deaths; the average number of weekly cases reported to the CDC was over 350,000 in the first weeks of 2023. In the State of Michigan, where this study was based, there have been a total of 3 million cases of COVID-19 reported and approximately 42,000 deaths (Centers for Disease Control and Prevention, 2023).

As a precautionary measure to curb the spread of the virus, the mobility of residents was severely restricted by lockdowns and other measures beginning in March, 2020 (Halford, Dixon, Farrell, Malleson, & Tilley, 2020, Mazerolle & Ransley, 2021). Schools and universities shifted to online learning; bars and restaurants were closed; and non-essential employees were forced to work remotely. Urban cores became ghost towns, as people were forced to remain home.

For environmental criminologists, stay-at-home orders, their subsequent impact on human mobility and everyday routine activities, and the social consequences of a nationwide pandemic have never been witnessed by modern scholars. Thus, the COVID-19 pandemic provides the unique opportunity to explore changes in patterns of criminal activity during an unprecedented time. Environmental criminology is a broad categorization for several related theories (i.e., rational choice, routine activities theory, and crime pattern theory) that are based on the assumption of a rational offender moving through time and space, seeking appropriate targets based on available information from their physical and social environment (Cornish & Clarke, 1986; Wortley & Mazerolle, 2008). Mobility is a critical component when examining opportunities for crime; when the movement of both offenders and potential victims dramatically changes, it follows that opportunities for crime would change as well.

Under RAT, crime occurs when a motivated offender encounters a suitable target that lacks guardianship (Clarke & Felson, 1993; Cohen & Felson, 1979; Felson, 1998; Felson, Jiang, & Xu, 2020). The location of the criminal event is important, as the offender has made the choice to attack a target in an area that is perceived to provide the greatest reward with the lowest level of apprehension risk. The purpose of this research is to compare the spatial locations of sexual assaults prior to, during, and after the COVID-19 mobility restrictions and resulting lifestyle changes that were enacted to prevent spread of the virus. Using data from the City of Detroit's Open Data Portal, the locations of reported sexual assaults that occurred in 3 distinct time frames were compared: Pre-COVID (11/25/2018–3/9/2020); COVID (3/10/2020–6/22/21); and Post-COVID (6/23/21–10/6/22).

The results of this study contribute to the body of knowledge in several important ways. First, relatively little research has been conducted on the geography of sexual assaults when compared to other forms of violent crime; this study adds to the nascent number of studies that have examined the importance of location in these crimes against persons. Second, risk terrain modeling (RTM) was used to identify potential risk factors for the occurrence of sexual assaults. While RTM has been used to study a number of violent crimes, this technique has been applied with less frequency to sexual assaults. A number of studies have included sexual assaults in an overall measure of crime, few have focused exclusively on sexual assault (Burgason, Drawve, Brown, & Eassey, 2017; Daley et al., 2016). Finally, this study contributes to our knowledge of variations in crime before, during, and after COVID 19 restrictions.1

2. Literature review

2.1. Theoretical perspectives: crime and place

Routine activities theory (RAT) is premised on the basic assumption of a rational offender carefully weighing the costs and benefits of his or her actions. RAT incorporates broad changes in contemporary society and the way these variations have impacted how we live our every lives. In order for crime to occur, three elements (commonly known as the crime triangle) need to converge in space and time: motivated offenders, suitable targets, and the absence of capable guardians (Cohen & Felson, 1979). RAT focuses attention on the patterns and movements of individuals as they live their daily lives; going to work, school, shopping, etc. Changes in routine activities can increase or decrease the opportunity for crimes to occur (Cohen & Felson, 1979; Leslie & Wilson, 2020).

Crime pattern theory further builds on RAT by incorporating an analysis of a city's geographic environment, such as land use patterns, street networks, and transportation systems. According to Brantingham and Brantingham, environmental criminology is the study of the spatio-temporal or locational dimension of crime. The focus is on when and where crimes will occur; what the movements are that bring the offender and the target together at the location of the crime; what is involved in the thought processes that lead to the selection of the crime location; and how targets and offenders are distributed spatially in urban, suburban, and rural settings (Brantingham & Brantingham, 1991; Brantingham & Brantingham, 1999; Brantingham & Brantingham, 2008).

Crime pattern theory directs attention at activity nodes and the pathways between them (Brantingham, Brantingham, & Andresen, 2016). Activity nodes are places where people travel to and from as part of their routine activities, such as shopping, work, school, or entertainment areas. Some activity nodes may become crime generators and crime attractors. Crime generators are locations that are frequently visited by large numbers of people, such as fast food restaurants, convenience stores, gas stations, etc. These locations support a variety of legitimate activities, but due to the sheer volume of people (both offenders and potential victims) that pass through these businesses, crime generators create more opportunities for crimes to occur. Conversely, crime attractors include such places as pawn brokers, check-cashing stores, methadone clinics, homeless shelters, and known locations for drug sales. A higher number of potential offenders or victims are likely to be attracted to these locations because of their purpose (Brantingham et al., 2016; Hart, Lersch, & Chataway, 2020).

2.2. Crime and location during COVID 19 mobility restrictions and lifestyle changes

In the examination of the impact of COVID-19 on levels of crime, many criminologists have focused on choice-based theories, which include elements of routine activities, crime pattern theory, and opportunity perspectives. Through this lens, it would not be surprising to see reductions in certain types of crimes, especially during pandemic lockdowns. Daytime residential burglaries and larcenies should decrease given that the stay-at-home orders increased the level of guardianship not only for the individual homeowner, but the entire neighborhood (Mohler et al., 2020; Stickle & Felson, 2020). Similarly, direct contact predatory crimes like robbery should also drop given the lack of potential targets and motivated offenders on the streets (Hodgkinson & Andresen, 2020a). Locations that would normally serve as attractors for motivated offenders and potential targets, such as bars, restaurants, concerts, schools, shopping malls, etc. were all shut down, which would again point to an expected decrease in crime (Brantingham & Brantingham, 1991; Brantingham & Brantingham, 2008; Hart et al., 2020).

2.3. The geography of sexual assaults

As the COVID-19 pandemic began to unfold, a number of researchers began to study the impact on a range of crimes, including street level drug activities (Abrams, 2020; Balmori de la Miyar, Hoehn-Velasco, & Silverio-Murillo, 2021); residential and commercial burglaries (Felson et al., 2020); “porch piracy,” which is the theft of delivered packages to homes (Stickle, 2020); shootings (Kim & Phillips, 2021); homicides and aggravated assaults (Campedelli, Aziani, & Favarin, 2020; Hodgkinson & Andresen, 2020b; Lersch & Hart, 2022) and other offenses. Incidents of violence against intimate partners, children, or other household members during COVID-19 have been examined in a number of studies (see, for example Barboza, Schiamberg, & Pachl, 2021; Lersch, 2022; Piquero, Jennings, Jemison, & Kaukinen, 2021; Piquero, Piquero, & Kurland, 2021; Shariati & Guerette, 2022). One crime that has been notably absent from this list is an in-depth examination of sexual assaults. While a number of researchers have explored the impact of the COVID-19 pandemic on the delivery of support services and health care to survivors of sexual assault (Engleton, Goodman-Williams, Javorka, Gregory, & Campbell, 2022; Munro-Kramer, Cannon, Scheiman, St Ivany, & Bailey, 2021; Wood et al., 2022), few studies have reported more than general trends, such as the decline in sexual assaults and other violent crimes during precautionary lockdowns (Bidstrup, Busch, Munkholm, & Banner, 2022; Payne, Morgan, & Piquero, 2022; Shen, Fu, & Noguchi, 2021).

In studies conducted prior to COVID-19 that explored the locations of sexual assaults, some have examined incidents in that occur in public settings (see, for example, Budd, Mancini, & Bierie, 2019), but most researchers conclude that the majority of these crimes occurred in private settings, such as the home of the victim or the offender (Colombino, Mercado, & Jeglic, 2009; Mogavero & Kennedy, 2017; Roberts, Donovan, & Durey, 2022). With respect to outdoor locations, public places that are “target rich,” “offender rich”,” and “guardian poor” pose especially high risk for sexual assault (Felson, de Melo, & Boivin, 2021). Therefore, attention must be directed at areas that are “target rich” and “offender rich,” such as public transit nodes, areas with a concentration of bars and clubs; entertainment districts; shopping malls and districts, convenience stores, etc. With respect to sexual assaults involving children, “youth-centric” locations, such as schools, churches, parks, and playgrounds may also offer opportunities for motivated offenders to select an attractive target (Budd et al., 2019). “Guardian poor” locations may include areas with few or no residences, limited visibility, and poor lighting, such as parking lots and garages (Roberts et al., 2022). In keeping with RAT and crime pattern theory, Felson and colleagues argued that when these three routine activity elements come together in time and space, there is a heightened level of risk of victimization.

2.4. Purpose of the present study

The purpose of this research is to compare the spatial locations of sexual assaults prior to, during, and after the COVID-19 mobility restrictions and lifestyle changes that were enacted during 2020. Using data from the City of Detroit's Open Data Portal, the locations of reported sexual assaults were grouped into 3 timeframes: Pre-COVID (11/25/2018–3/9/2020); COVID (3/10/2020–6/22/21); and Post-COVID (6/23/21–10/6/22). The following research questions were explored:

  • 1.

    Were there differences in the spatial distribution Pre-COVID, COVID, and Post-COVID time frames?

  • 2.

    Following the tenets of environmental criminology, were there identifiable spatial risk factors associated with target rich, offender rich, or guardianship poor locations that increased the risk of sexual assault?

  • 3.

    Did the risk factors change when comparing the Pre-COVID, COVID, and Post-COVID locations?

2.5. The setting: Detroit, Michigan, USA

The City of Detroit is located within Wayne County in Southeastern Michigan. The population in 2020 was 639,111 persons. The median household income was $32,498 and 33.2% of the City's residents lived in poverty. The majority of residents were Black or African American (77.1%), with 14.4% White, and the remaining classified as other. Most Detroit residents were not Hispanic or Latino (92.3%) (United States Census Bureau, 2022).

There are a number of characteristics of the City of Detroit that make it unique. First and foremost is the catastrophic reduction in population. The population of the city dropped from nearly two million residents in 1950 to its current level of under 700,000 residents (Binelli, 2013). This mass exodus from the city resulted in a second unique attribute: the proliferation of abandoned and demolished structures. At the peak of the City's decline, nearly 100,000 structures stood empty and abandoned. The City encompasses an area of 139 mile2 with nearly 40 mile2 of vacant land due to demolitions and property abandonment (Binelli, 2013). Finally, Detroit is unique due to its consistently high crime rate. Detroit is one of the most violent cities in the world. According to the FBI, the violent crime rate in Detroit was 1965 violent crimes per 100,000 people, which was much higher than the national level of 366.7 violent offenses per 100,000 (Federal Bureau of Investigation, 2019).

The City of Detroit completely encompasses two distinct municipalities that support their own police departments: Hamtramck and Highland Park. Comparable data were unavailable for these two cities, which were excluded from the analyses. A boundary file was downloaded from the U.S. Census Bureau for the City of Detroit. This shapefile was projected to Mercator Auxiliary Sphere with a geographic coordinate system of WGS 1984 using ArcGIS Pro 3.1.

2.6. Detroit and COVID-19

The City of Detroit was thrust into the national spotlight on April 5th, 2020, when the metropolitan area was identified as an outbreak hotspot during a COVID-19 Task Force briefing. At that time, Detroit reported the third highest number of cases in the US (WXYZ, 2020, April 6).2 The Detroit metropolitan area has been described as a “hardest hit place”, ranked 7th highest in the US with respect to the rate of cumulative confirmed deaths, behind New York City and New Orleans (New York Times, 2020, May 15).

The unique characteristics of the City of Detroit make the area particularly vulnerable to the spread of COVID-19. Along with the high poverty rate, residents report higher levels of obesity, diabetes, hypertension, and a lack of access to health insurance and reliable transportation (Abdalla, 2020, April 3). Additionally, Detroit is home to one of the largest African American populations in the US. Nationally, the COVID-19 rate is disproportionately higher among African Americans. In the state of Michigan, African Americans account for 40% of the deaths due to the coronavirus, but only 14% of the statewide population is African American (Shay, 2020, April 10).

On March 10th, 2020, Michigan Governor Gretchen Whitmer declared a state of emergency after 2 cases of COVID-19 were reported. Shortly thereafter, all K-12 schools were closed for in-person classes (3/12/2020), recreational venues were shut down, and a ban on dine-in eating was implemented (3/16/2020). The first confirmed death due to COVID-19 was on 3/18/2020, and a stay-at-home order was issued on 3/23/2020 (University of Michigan School of Public Health, 2022). Governor Whitmer's response has been described as early and aggressive, and comparatively, the state of Michigan ranked 13/50 states for having the most COVID-19 restrictions (McCann, 2021). On June 22, 2021, all COVID-related interventions, including restrictions on indoor capacities, mask mandates, and social distancing requirements were lifted in Michigan. All bars, restaurants, movie theaters and gyms returned to 100% indoor service (Livengood, 2021). This ended 15 months of restrictions that began on 3/10/2020, when the Governor first began to discourage travel and limit gatherings of more than 100 persons (University of Michigan School of Public Health, 2022).

For purposes of the present analysis, the COVID time frame was defined as beginning on 3/10/2020 and ending on 6/22/2021,3 a period of 470 days. The Pre-COVID time frame was defined as the 470 days prior to 3/10/2020 and the Post-COVID time frame was operationalized as the 470 days after 6/22/21. The equal distribution of the data points before, during, and after the time of COVID-19 restrictions increases statistical power (Bernal, Cummins, & Gasparrini, 2017; Demir & Cassino, 2022) and allows for a stronger comparison of the outcomes (Madero-Hernandez, Deryol, Ozer, & Engel, 2017), which allows for the detection of small changes, thereby increasing the robustness of the results (see Bernal et al., 2017).

2.7. Analytical plan: hot spot analysis and risk terrain modeling

To explore the research questions, first optimized hot spot analysis was used. This tool uses the Getis-Ord Gi* statistic, which identifies areas where a feature has a high (or low) number of incidents and is surrounded by other features that similarly have high (or low) values (Lentz, 2009). The optimized hot spot analysis tool mines the data in order to obtain parameters that will yield optimal hot spot results (Esri, 2018). The tool aggregates the point data, identifies the correct scale of analysis, and automatically corrects for multiple testing and spatial dependence. Statistically significant results indicate the locations where point clusters are unusually intense or sparse.

Second, risk terrain modeling was used to identify high risk locations for sexual assaults. Risk Terrain Modeling (RTM) is a relatively new analytical technique developed by Joel Caplan and Leslie Kennedy as they studied the locations of shootings in Irvington, New Jersey. Using geographic information systems (GIS) technology, spatial relationships were examined between the locations of shootings and other factors identified as potentially “risky” for the occurrence of shootings: known gang members' addresses; the locations of drug arrests; and high-risk infrastructure, which was defined as the locations of liquor stores, bars, strip clubs and fast food restaurants. The authors then employed these risk factors to develop maps to assist law enforcement personnel in the prediction of the locations of future shooting incidents, develop plans for intervention, and assess the effectiveness of the response (Caplan & Kennedy, 2013; Caplan, Kennedy, & Miller, 2011; Caplan, Kennedy, & Piza, 2013).

The underlying theoretical basis for RTM is based on the notion that crime is not randomly distributed through time and place: Some locations are “riskier” than others for various forms of victimization. RTM is heavily influenced by environmental criminologists who have focused on the geography of crime, including early Chicago School theorists, routine activities and opportunity theory, and the Brantinghams' notion of crime generators, attractors, and protective factors (for a full discussion of the principles of environmental criminology, please see Brantingham & Brantingham, 1991 or Wortley & Mazerolle, 2008).

On a basic level, the general idea behind RTM is to create a grid of cells (say, a 500 by 500 m raster grid) that covers the entire study area. To measure the influence of proximity, for each factor of interest a value of 1 is assigned to cells in which the risk is present (such as the area one block around a bar) while a 0 is assigned to a cell where the risk is not present (i.e., no bar is present). A risk layer can also be created the measures the influence of density, such as an entertainment area with a high concentration of bars and restaurants. An individual risk layer is created for each factor in the analysis. The different risk layers are ultimately combined into a final Risk Terrain Map in which the cell values for each corresponding 500 × 500 area are summed together.

3. Method

3.1. Reported sexual assaults

The City of Detroit maintains a robust open data portal that provides access to maps and datasets focused on a variety of government services, including public safety, health, transportation, and land use (City of Detroit, 2023). Launched in 2015, the portal was funded through a grant from the Socrata Foundation in an effort to assist in redevelopment efforts (Shueh, 2015, March 12). The Open Data Portal was the source for many of the datasets used in this analysis, as can be seen in Appendix A.

One of the datasets maintained in the Open Data Portal contains reported criminal offenses that have occurred in the City of Detroit. The data are extracted from the Detroit Police Department's (DPD) records management system and are updated when the DPD sends its offense reports to the Michigan Incident Crime Reporting or the National Incident Based Reporting Systems. Reported sexual assaults were selected from this database. According to Michigan Penal Code Section 750.520, “Sexual assault is any form of unwanted sexual contact obtained without consent and/or obtained through the use of force, threat of force, intimidation, or coercion (FindLaw, 2018).” Michigan law further defines three degrees of felonious sexual assault; misdemeanor 4th degree sexual contact was not included in this analysis (for a full definition of 1st, 2nd, and 3rd degree sexual assault please see Appendix B).

For the Pre-COVID period (defined as 11/25/2018–3/9/2020) there were 1040 sexual assaults or 2.21 assaults per day. In the COVID period (defined as 3/10/2020–6/22/2021) there were 887 sexual assaults, or 1.89 assaults per day. Finally, in the 470 days of the Post-COVID period, there were 833 sexual assaults, or 1.77 per day.

3.2. Risk factors

A total of 17 risk factors were tested using the RTMDx Diagnostic software: banks / check cashing businesses (n = 161); beauty salons, day spas, and nail salons (n = 362); blight complaints (n = 84,025); casinos (n = 3); dangerous drugs arrests (n = 6157); demolitions (25,377); active liquor points of sale (n = 1026); laundromats (n = 82); parking lots and garages (n = 55); parks (n = 338); public transit stops (n = 10,939); recreation centers (n = 11); schools (n = 546); social service locations (n = 24); supermarkets (n = 307); major sports venues (n = 4); and worship centers (n = 231). The RTM Diagnostics software conducts tests for the spatial influence of both proximity and density of the risk factors. That is, the software tests to see if being within a distance of two blocks of a parking lot puts one at higher risk for a sexual assault (proximity using Euclidian distance) or if being in an area where there is a high concentration of drug arrests is risky (density, as measured by kernel density). “High density” blocks are defined as those areas that are 2 standard deviations above the mean density value.

The street grid system for the City of Detroit results in average blocks of less than 2.7 acres, or a block length of approximately 125 m (Rayle, 2011). As the spatial influence of risk factors is normally just a few blocks (Caplan et al., 2013), the default value of 3 block lengths (or 375 m) was used. This implied that spatial influence was tested at up to 1 block, up to 2 blocks, and up to 3 blocks. As part of the analytical process a raster cell size of 50 m by 50 m was created for testing.

The process began by building an elastic net penalized regression model assuming a Poisson distribution of events.4 Through a cross validation process, risk factors with non-zero coefficients were identified as potentially of use. These remaining variables were then used in a bidirectional step-wise regression to build a model that optimized the Akaike Information Criterion (AIC), a score that balances the fit of the model against its complexity. The stepwise regression process was conducted for both Poisson and Negative Binomial distributions with the lowest AIC score used to identify the appropriate distribution (for a more detailed discussion of the testing process, please see (Caplan et al., 2013; Heffner, 2013). The coefficients in the final models were rescaled to produce relative risk values (RRVs), weighting each factor relative to one another. Finally, relative risk scores were calculated for each cell in the study area (Barnum, Caplan, Kennedy, & Piza, 2017).

4. Results

The first research question in this analysis focused on the geographic distribution of sexual assaults in the City of Detroit. Specifically, were there differences in the spatial distribution of sexual assaults during the Pre-COVID, COVID, and Post-COVID timeframes? The maps below display the results of optimized hot spot analysis, which show statistically significant hot and cold spots (p < .05) for sexual assaults (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Optimized hot spot analysis results.

When comparing the maps, the hot spots were more intense and covered a larger area during the Pre-COVID time frame. This is especially true in the core downtown area. During the COVID time frame, no statistically significant hot spots were found in the downtown area. In the Post-COVID time frame, one can see the development of significant hot spots in the central downtown area once again. The locations of the cold spots (in blue) remained relatively consistent across the three time frames.

4.1. Risk terrain modeling results

The remaining research questions focused on the risk terrain modeling results; specifically, were there identifiable spatial risk factors associated with target rich, offender rich, or guardianship poor locations that increased the risk of sexual assault? And did these risk factors change when comparing the Pre-COVID, COVID, and Post-COVID timeframes?

4.2. Pre-COVID timeframe

The first risk terrain model was a Negative Binomial Type II model that included 4 risk factors and yielded an AIC score of 11,295.21. A Relative Risk Score (RRS) was assigned to each place in the study area, ranging from 1 for the lowest risk to 23.07 for the highest risk place. The RRS values allow for easy comparison of places; An RRS value of 10 has an expected rate of sexual assault that is 2 times higher than a place with an RRS of 5. The analysis was based on a place size of 125 m, resulting in a total number of 149,718 places. The mean RRS was 3.36 with a standard deviation of 3.05. The number of places greater than 2 standard deviations from the mean was 6560, or 4.38% of the City of Detroit. Of the 17 risk factors tested, 4 were included in the final model, as noted below.

4.3. COVID-19 time frame

The second risk terrain model was a Negative Binomial Type II model that included 6 risk factors and yielded an AIC score of 9862.76. The coefficients derived in the final model were used to manually construct a risk terrain map to display Relative Risk Scores (RRS). A Relative Risk Score (RRS) was assigned to each place in the study area, ranging from 1 for the lowest risk to 867.55 for the highest risk place. The analysis was based on a place size of 125 m, resulting in a total number of 149,718 places. The mean RRS was 17.75 with a standard deviation of 18.13. The number of places greater than 2 standard deviations from the mean was 4006, or 2.68% of the City of Detroit. Of the 17 risk factors tested, 6 were included in the final model, as noted in Table 1 . Places affected by a risk factor with an RRV of 12 are 4 times as risky compared to places affected by a risk factor with a RRV of 3.

Table 1.

Risk factors and their relative risk values for the pre-COVID, COVID, and post-COVID time frames.

 Pre-COVID
 COVID
 Post-COVID
Risk Factor Proximity/Density Spatial Influence RRV Proximity/Density Spatial Influence RRV Proximity/Density Spatial Influence RRV
Casinos Proximity 125 m 11.417
Blight Complaints Proximity 125 m 2.918 Proximity 250 m 9.072 Proximity 125 m 2.223
Public Transit Stops Density 125 m 2.335 Density 125 m 2.853 Density 125 m 2.646
Drug Locations Density 125 m 2.226 Density 125 m 2.204 Proximity 125 m 1.925
Liquor Outlets Density 125 m 1.494 Proximity 375 m 1.332 Density 125 m 1.521
Demolitions Proximity 125 m 1.412
Schools Proximity 375 m 1.317
Open in a new tab

Gov. Gretchen Whitmer's administration ordered Detroit casinos closed on March 16, 2020. They stayed closed until August 5, 2020, but then they were ordered closed again on November 18, 2020. They reopened on December 21, 2020, at limited capacity.

4.4. Post-COVID timeframe

The final risk terrain model was a Negative Binomial Type II model that included 5 risk factors and yielded an AIC score of 9476.42. The coefficients derived in the final model were used to manually construct a risk terrain map to display Relative Risk Scores (RRS). A Relative Risk Score (RRS) was assigned to each place in the study area, ranging from 1 for the lowest risk to 22.69 for the highest risk place. The analysis was based on a place size of 125 m, resulting in a total number of 149,718 places. The mean RRS was 3.85 with a standard deviation of 3.54. The number of places greater than 2 standard deviations from the mean was 10,310, or 6.89% of the City of Detroit. Of the 17 risk factors tested, 5 were included in the final model, as noted in Table 1.

5. Discussion

The purpose of this research was to explore potential changes in serious sexual assaults that occurred before, during, and after COVID 19 restrictions were mandated. Overall, there were fewer sexual assaults per day in the COVID time frame. For the Pre-COVID period (defined as 11/25/2018–3/9/2020) there were 1040 sexual assaults or 2.21 assaults per day. In the COVID period (defined as 3/10/2020–6/22/21) there were 887 sexual assaults, or 1.89 assaults per day. Finally, in the 470 days of the Post-COVID period (defined as 6/23/21–10/6/22), there were 833 sexual assaults, or 1.77 per day. This finding was consistent with the results obtained from other studies that reported a decrease in the level of other types of violent crime in the days and months following COVID-19 (Campedelli et al., 2020; Hodgkinson & Andresen, 2020b; Kim & Phillips, 2021; Shariati & Guerette, 2022).

With respect to location, generally speaking, the results of the Optimized Hot Spot Analysis indicated that hot spot areas within the City of Detroit were larger and more intense in the Pre-COVID 19 time frame. This was especially pronounced in the area around downtown. During the Pre-COVID time frame, approximately 4.38% of the downtown area was a hot spot (p < .05); comparatively, only 2.68% of the downtown area was a hot spot during COVID, but increased to 6.89% after restrictions were lifted. This finding was consistent with routine activities theory, where changes in daily routines can increase or decrease the opportunity for crimes to occur. The downtown area contains the business and financial districts, sport venues, casinos and entertainment areas, as well as approximately 11,000 residents (Point2Homes, 2020). Additionally, nearly 160,000 commuters made their way to Detroit for work from the surrounding suburbs during the Pre-COVID timeframe (Abello, 2017). When the COVID mobility restrictions were initiated, the number of motivated offenders and targets decreased, resulting in a decline in both the number of sexual assaults per day and a smaller area where the crimes were concentrated.

The results of the RTM analyses revealed some interesting trends. For example, the proximity of casinos and demolitions sites were the only risk factor whose influence on crime was limited only to the COVID-19 time frame. For casinos, this was likely due to their resistance to closing and push to reopen quickly in response to statewide mandates restricting human interaction. Gov. Gretchen Whitmer's administration ordered Detroit casinos closed on March 16, 2020. Casinos stayed closed until August 5, 2020, when they attempted to reopen. They were ordered closed again on November 18, 2020. They reopened and stayed open on December 21, 2020, at limited capacity (Craig, 2020). Their influence on crime patterns during the COVID period (i.e., 3/10/2020–6/22/2021) is not surprising, given that Detroit casinos were open during much of this time, providing one of the few large entertainment options in the city where people could interact.

As for demolitions sites, vacant structures and land parcels impact the level of guardianship and informal social control. Safe neighborhoods are characterized by natural proprietors keeping their “eyes upon the street” (Jacobs, 1962). If a rational offender perceived a strong sense of community, apprehension and detection would be more likely since local residents would notice their presence and step in if they attempted to commit a crime (Taylor & Gottfredson, 1986). When neighborhoods deteriorate and populations decline, the eyes on the street are lost, resulting in poor guardianship and higher risk of sexual assaults. Current findings suggest this was an especially acute during COVID-19 in areas proximate to demolitions sites.

Findings from the present study that are associated with other risk factors are also noteworthy, including blight complaints As argued by Felson et al. (2021), public places that are target rich, offender rich, and guardian poor pose especially high risk for sexual assault. Blight complaints were found to be the risk factor with the highest relative risk value during the COVID period. Blight complaints can have a direct impact on guardianship. Wesley Skogan has written on the interaction between disorder, fear of crime, and community oversight. Physical signs of disorder, such as vandalism, graffiti, junk and trash in vacant lots, abandoned cars, etc., contribute to feelings of insecurity and fear of crime. People who are experiencing heightened levels of fear withdraw from their neighborhoods. They no longer feel comfortable walking the streets after dark, nor do they get involved when they see behavior that violates the informal social controls of the neighborhood. Fear of crime reduces the spatial radius, or the territory that an individual feels some sense of responsibility to defend and protect. Untended areas then become prime areas for crimes to occur (Skogan, 1986, Skogan, 1990).

Public transit stops, locations of serious drug offenses, and liquor outlets were also found to be important risk factors. Their influence on crime patterns were relatively more consistent across all three COVID time frames examined in the current study. These places would fall into what Felson et al. (2021) described as “offender rich” and “target rich.” For example, public transit stops are frequented by both offenders and potential victims as they move through their daily activities. Public transit remained operational during the pandemic to ensure that essential workers, such as nurses, grocery clerks, janitors, etc., could get to their place of employment. Even though ridership declined during the COVID timeframe (Transportation Riders United, 2020), these transportation nodes remained an important risk factor. Bus stops can be particularly risky places for several reasons. Vulnerable targets, such as women and children, may walk alone from their home to a bus stop. Further, riders on busses and other public transportation can be strangers which can lead to chance encounters between offenders and victims (Brantingham & Brantingham, 2008). Similarly, liquor outlets could include convenience stores, restaurants, bars, or nightclubs, examples s of “offender rich” and “target rich” locations as described by Felson et al. (2021).

Finally, patterns observed in the current study associated with schools were interesting. Schools were given as one example of a youth-centric location by Budd et al., 2019, in a study examining high risk areas for sexual assaults involving children. In the present analysis, areas that were within 3 blocks of a school posed heightened risk, but only during the post-COVID time frame. These findings suggest unique changes in everyday routine activities after restrictions were lifted in Detroit that changed the opportunity structure of rape and sexual assaults near schools.

5.1. Limitations of the present study

These results are based on sexual assaults that have been reported to the police, and based on the investigation by law enforcement, it was determined that a felonious sexual assault had occurred. The assaults used in this analysis have gone through several filters, especially with respect to under-reporting of sex-based offenses. Self-report studies of victimization have found that nearly 80% of sexual assaults are not reported to the police (Kimble & Chattier, 2018). The rate of reporting varies based on the age of the victim, the relationship between the victim and the perpetrator, and other factors. For example, 86% of adolescent sexual assaults are not reported; a similar percentage has been found among college students. When the perpetrator is known to the victim (i.e., former husbands or boyfriends) the rate of reporting is about 23%; for friends and acquaintances the rate increases to 39%. If the perpetrator is a stranger, only about half of the crimes are reported to the police (Office for Victims of Crime, 2005). The true number of sexual assaults cannot be known. Additionally, no detail was provided in the sexual assault data regarding the age, race, gender, or any potential relationship between the victim and offender. Only the XY coordinates of the reported sexual assault were provided.

Another limitation of the present study involves the accuracy of risk factor locations. That is, while the bars may have an active liquor license (and therefore be included as a risk factor), it cannot be known if / when the bar opened and closed throughout each COVID time frame. Similarly, the bus routes were modified throughout the COVID time frame as new routes were added and other closed. One of the critical assumptions of RTM concerns the accuracy of the risk factor locations. The results and conclusions should be interpreted with caution given the fluid nature of the data sources.

Finally, while the results provided important insight into spatial risk factors for sexual assault before and during the COVID mobility restrictions, the findings may not apply in other jurisdictions given the unique traits of the City of Detroit as well as the relatively strong restrictions imposed in the State of Michigan. It would be of interest to replicate this study in other cities and states to ascertain the generalizability of the results.

Footnotes

1

In the present study, ‘restrictions’ were broadly defined to include not only mandated stay at home orders, but also other limitations on routine activities, such as size restrictions on gatherings of individuals, reduced capacities at restaurant, bars, and other service locations, school closures, and other disruptive proactive measures enacted to reduce the spread of the COVID-19 virus.

2

While it is recognized that references to primary sources (i.e., the City of Detroit or Wayne County Public Health) were preferable, in some cases the benchmarks for the impact of the COVID virus were found in media and/or secondary reports.

3

It is acknowledged that different dates could have been selected for marking the beginning and end of the COVID-19 time frame. As the purpose of this manuscript focused more broadly on changes to routine activities than just mobility restrictions (i.e., lockdowns or stay at home mandates), the 3/10/2020–6/22/21 time frame was justified. For example, all K-12 schools were closed for in-person learning on 3/12/2020, which was the first major disruption for routine activities.

4

A poison distribution was deemed appropriate as count data was used. Alternatively, rates could have used to consider possible size differences in the volume of sexual assaults. It would be difficult, if not impossible, to determine an accurate population estimate for the calculation of rates given the movement (or lack thereof) of individuals. It has been argued that up to 80% of the daily activity space of individuals occurs outside of the area perceived to be their home neighborhood (Helbich, 2018). The use of rates may be misleading due to the fluid changes to underlying population especially during the Pre-COVID timeframe, especially in the downtown areas as commuters go to their workplaces.

Appendix A. Risk factors

Risk Factor Source Date Accessed Description
Banks / Check
Cashing		 InfoGroup / Esri Business Analyst 8/22/16 Banks, check cashing services, payday loans, etc.
Blight Violations Open Data Portal 10/28/22 Complaints against property owners of commercial and residential properties from 1/1/17–4/15/19
Casinos Google Maps 11/2/22 Point shapefile created from address data
Drug Arrests Open Data Portal: RMS Crime Incidents 10/24/22 Includes incidents categorized as “Dangerous drugs” from 1/1/17–4/15/19; cocaine-sell, crack-sell, heroin-sell/manufacture, marijuana-sell; synthetic narcotic-sell
Demolitions Open Data Portal: Property Parcels Demolitions 10/27/22 Includes all commercial and residential demolitions from 1/1/2014–10/14/22
Liquor Licenses Michigan Liquor Control Commission 5/3/19 Active liquor licenses; includes bars, restaurants, liquor / convenience stores
Parking lots / garages InfoGroup / Esri Business Analyst 4/4/2017 Businesses with NAICS Code 7225117
Parks Open Data Portal: Parks 2016 5/9/2019 Polygon shapefile downloaded from source
Public Transit Stops Open Data Portal: Transportation 10/24/22 Bus stops, SMART Bus Stops, Q Line Stops; File updated 11/16/2019
Recreation Centers Open Data Portal: Recreation Centers 5/9/2019 Point /address file downloaded from source
Schools Open Data Portal : 2018/2019 Schools 10/25/22 Includes all active schools (public, private, and charter)
Social Service Locations InfoGroup / Esri Business Analyst 8/22/16 Includes homeless shelters, food banks, and substance abuse centers
Sports Venues Esri Imagery Basemap 10/26/22 Polygon shapefile created from areal basemap imagery
Worship Centers Google Maps 5/11/2019 Point shapefile created from address data; Includes churches, synagogues and mosques
Open in a new tab

Appendix B. Degrees of sexual assaults Michigan penal code section 750.520

Sexual Assault: Sexual assault is any form of unwanted sexual contact obtained without consent and/or obtained through the use of force, threat of force, intimidation, or coercion.

1st Degree: A sexual act involving penetration and any of the following:

  • •

    Victim is under 13 years old;

  • •

    Victim is 13–15 years old and is a blood affiliation to the defendant, lives in the defendant's household, or the defendant is in an authority position to the victim;

  • •

    Multiple actors are involved and force/coercion was used to accomplish the sexual penetration or the victim is incapacitated (physically helpless, mentally incapacitated or mentally defective) weapon involved;

  • •

    Personal injury and force / coercion;

  • •

    Personal injury and victim incapacitated (Unable to consent due to age, mentally challenged or due to intoxication, date rape drug, etc.); or

  • •

    Defendant is in the process of committing another felony.

2nd Degree: Sexual contact (No penetration) with the genital area, groin, inner thigh, buttock or break, and any f the circumstances listed for 1st Degree Criminal Sexual Contact.

3rd Degree: Sexual Penetration and any of the following:

  • •

    Victim is 13–15 years old;

  • •

    Force or coercion; or

  • •

    Victim is incapacitated (unable to consent due to age, mental challenges, intoxication, date rap drug, etc).

References

  1. Abdalla J. Al Jazeera News; 2020, April 3. Detroit especially vulnerable to COVID-19 due to poverty, health.https://www.aljazeera.com/news/2020/04/detroit-vulnerable-covid-19-due-poverty-health-200403140214218.html Retrieved May 16, 2020, from. [Google Scholar]
  2. Abello O.P. These Detroit commuting numbers show stark inequality. 2017. https://nextcity.org/urbanist-news/new-report-detroit-jobs-numbers-pay-commute Retrieved January 13 from.
  3. Abrams D.S. COVID and crime: An early empirical look [working paper] 2020. https://ssrn.com/abstract=3674032 [DOI] [PMC free article] [PubMed]
  4. Balmori de la Miyar J.R., Hoehn-Velasco L., Silverio-Murillo A. Druglords don’t stay at home: COVID-19 pandemic and crime patterns in Mexico City [article] Journal of Criminal Justice. 2021;72 doi: 10.1016/j.jcrimjus.2020.101745. N.PAG-N.PAG. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barboza G.E., Schiamberg L.B., Pachl L. A spatiotemporal analysis of the impact of COVID-19 on child abuse and neglect in the city of Los Angeles, California. Child Abuse & Neglect. 2021;116 doi: 10.1016/j.chiabu.2020.104740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Barnum J., Caplan J., Kennedy L., Piza E.L. The crime kaleidoscope: A cross-jurisdictional analysis of place features and crime in three urban environments. Applied Geography. 2017;79:203–211. doi: 10.1016/j.apgeog.2016.12.011. [DOI] [Google Scholar]
  7. Bernal J.L., Cummins S., Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: A tutorial. International Journal of Epidemiology. 2017;46(1):348–355. doi: 10.1093/ije/dyw098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bidstrup J.E., Busch J.R., Munkholm J., Banner J. Impact of the Covid-19 lockdown on sexual assault cases in eastern Denmark – A retrospective clinical forensic study. Forensic Science, Medicine, and Pathology. 2022;18(2):125–132. doi: 10.1007/s12024-021-00455-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Binelli M. Picador; 2013. Detroit city is the place to be. [Google Scholar]
  10. Brantingham P.J., Brantingham P.L. 2nd ed. Waveland Press; 1991. Environmental Criminology. [Google Scholar]
  11. Brantingham P.J., Brantingham P.L. In: Environmental criminology and crime analysis. Wortley R., Mazerolle L., editors. Willan Publishing; 2008. Crime pattern theory. [Google Scholar]
  12. Brantingham P.J., Brantingham P.L., Andresen M.A. In: Environmental criminology and crime analysis. 2nd ed. Wortley R., Townsley M., editors. Routledge; 2016. The geometry of crime and crime pattern theory; pp. 119–136. [Google Scholar]
  13. Brantingham P.L., Brantingham P.J. Vol. 8. 1999. A theoretical model of crime hot spot generation. Studies on crime and crime prevention; pp. 7–26. [Google Scholar]
  14. Budd K.M., Mancini C., Bierie D.M. Parks, playgrounds, and incidents of sexual assault. Sexual Abuse. 2019;31(5):580–606. doi: 10.1177/1079063218797712. [DOI] [PubMed] [Google Scholar]
  15. Burgason K., Drawve G., Brown T.C., Eassey J. Close only counts in alcohol and violence: Controlling violence near late-night alcohol establishments using a routine activities approach. Journal of Criminal Justice. 2017;50:62–68. doi: 10.1016/j.jcrimjus.2017.04.004. [DOI] [Google Scholar]
  16. Campedelli G.M., Aziani A., Favarin S. Exploring the immediate effects of COVID-19 containment policies on crime: An empirical analysis of the short-term aftermath in Los Angeles. American Journal of Criminal Justice. 2020 doi: 10.1007/s12103-020-09578-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Caplan J., Kennedy L. Rutgers Center on Public Security; 2013. Risk terrain modeling diagnostics utility. [Google Scholar]
  18. Caplan J., Kennedy L., Piza E.L. Rutgers Center on Public Security; 2013. Risk terrain modeling diagnostics utility user manual version 1.0. [Google Scholar]
  19. Caplan J.M., Kennedy L., Miller J. Risk terrain modeling: Brokering criminological theory and GIS methods for crime forecasting. Justice Quarterly. 2011;28(2):361–381. doi: 10.1080/07418825.2010.486037. [DOI] [Google Scholar]
  20. Centers for Disease Control and Prevention CDC Museum COVID-19 Timeline. 2022. https://www.cdc.gov/museum/timeline/covid19.html#Mid-2020 Retrieved February 9 from.
  21. Centers for Disease Control and Prevention . US Department of Health and Human Services; 2023. COVID Data Tracker.https://covid.cdc.gov/covid-data-tracker Retrieved February 9 from. [Google Scholar]
  22. City of Detroit Detroit's open data portal. 2023. https://data.detroitmi.gov/
  23. Clarke R., Felson M. Transaction Publishers; 1993. Routine activity and rational choice: Advances in criminological theory volume 5. [Google Scholar]
  24. Cohen L., Felson M. Social change and crime rate trends: A routine activity approach. American Sociological Review. 1979;44:588–608. [Google Scholar]
  25. Colombino N., Mercado C., Jeglic E. Situational aspects of sexual offending: Implications for residence restriction laws. Justice Research and Policy. 2009;11(1–2):27–43. doi: 10.3818/JRP.11.2009.27. [DOI] [Google Scholar]
  26. Cornish D., Clarke R. Springer-Verlag; 1986. The reasoning criminal. [Google Scholar]
  27. Craig K. WXYZ Detroit; 2020. 2 Detroit casinos to open to the public today; Here’s what to expect. (Retrieved January 13 from) [Google Scholar]
  28. Daley D., Bachmann M., Bachmann B., Pedigo C., Bui M., Coffman J. Risk terrain modeling predicts child maltreatment. Child Abuse & Neglect. 2016;62:29–38. doi: 10.1016/j.chiabu.2016.09.014. [DOI] [PubMed] [Google Scholar]
  29. Demir M., Cassino P.P. The effect of COVID-19 on police activities: Traffic stops, arrests and use of force. Criminal Justice Review. 2022 doi: 10.1177/07340168221139356. [DOI] [Google Scholar]
  30. Engleton J., Goodman-Williams R., Javorka M., Gregory K., Campbell R. Sexual assault survivors’ engagement with advocacy services during the COVID-19 pandemic. Journal of Community Psychology. 2022;50(6):2644–2658. doi: 10.1002/jcop.22819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Esri How optimized hot spot analysis works. 2018. https://pro.arcgis.com/en/pro-app/tool-reference/spatial-statistics/how-optimized-hot-spot-analysis-works.htm Retrieved May 27, 2019 from.
  32. Federal Bureau of Investigation . U.S. Department of Justice; 2019. Crime in the United States.https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-2019 [Google Scholar]
  33. Felson M. Pine Forge Press; 1998. Crime and everyday life. [Google Scholar]
  34. Felson M., de Melo S.N., Boivin R. Risk of outdoor rape and proximity to bus stops, bars, and residences. Violence and Victims. 2021;36(6):723–738. doi: 10.1891/VV-D-20-00074. [DOI] [PubMed] [Google Scholar]
  35. Felson M., Jiang S., Xu Y. Routine activity effects of the COVID-19 pandemic on burglary in Detroit. Crime Science. 2020;9 doi: 10.1186/s40163-020-00120-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. FindLaw Michigan Rape Laws. 2018. https://www.findlaw.com/state/michigan-law/michigan-rape-laws.html
  37. Halford E., Dixon A., Farrell G., Malleson N., Tilley N. Crime and coronavirus: Social distancing, lockdown, and the mobility elasticity of crime [article] Crime Science. 2020;9(1):1–12. doi: 10.1186/s40163-020-00121-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Hart T., Lersch K.M., Chataway M. 5th ed. Carolina Academic Press; 2020. Space, time & crime. [Google Scholar]
  39. Heffner J. In: Rutgers center on public security. Caplan J., Kennedy L., editors. 2013. Risk terrain modeling diagnostics utility user manual version 1.0. [Google Scholar]
  40. Hodgkinson T., Andresen M.A. Show me a man or a woman alone and I’ll show you a saint: Changes in the frequency of criminal incidents during the COVID-19 pandemic. Journal of Criminal Justice. 2020;69 doi: 10.1016/j.jcrimjus.2020.101706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Hodgkinson T., Andresen M.A. Show me a man or a woman alone and I’ll show you a saint: Changes in the frequency of criminal incidents during the COVID-19 pandemic [article] Journal of Criminal Justice. 2020;69 doi: 10.1016/j.jcrimjus.2020.101706. N.PAG-N.PAG. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Jacobs J. Cape; 1962. The death and life of great American cities. [Google Scholar]
  43. Kim D.-Y., Phillips S.W. When COVID-19 and guns meet: A rise in shootings [article] Journal of Criminal Justice. 2021;73 doi: 10.1016/j.jcrimjus.2021.101783. N.PAG-N.PAG. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Kimble C., Chattier I. Sexual assault remains dramatically underreported. 2018. https://www.brennancenter.org/our-work/analysis-opinion/sexual-assault-remains-dramatically-underreported
  45. Lentz J. Spatial autocorrelation statistics. 2009. http://jenniferalentz.info/Teaching/StudyGuides/SpatialAutocorrelation.pdf Retrieved July 15, 2019 from.
  46. Lersch K. International journal of comparative and applied criminal justice; Florida, USA: 2022. An exploratory spatiotemporal analysis of domestic disturbance calls for service in 2020: The case of Tampa. [DOI] [Google Scholar]
  47. Lersch K., Hart T.C. COVID-19, violent crime, and domestic violence: An exploratory analysis. International Criminology. 2022;2(3):1–14. doi: 10.1007/s43576-022-00049-y. [DOI] [Google Scholar]
  48. Leslie E., Wilson R. Sheltering in place and domestic violence: Evidencee from calls for service during COVID-19. Journal of Public Economics. 2020;189 doi: 10.1016/j.jpubeco.2020.104241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Livengood C. Michigan lifting all COVID-19 capacity, mask restrictions June 22. 2021. https://www.detroitchamber.com/michigan-lifting-all-covid-19-capacity-mask-restrictions-june-22/ Retrieved February 14 from.
  50. Madero-Hernandez A., Deryol R., Ozer M., Engel S. Examining the impact of early childhood school investments on neighborhood crime. Justice Quarterly. 2017;34(5):759–787. [Google Scholar]
  51. Mazerolle L., Ransley J. Policing health regulations in democratic societies: A focus on COVID-19 challenges and opportunities in Australia. International Journal of Comparative and Applied Criminal Justice. 2021;45(3):315–327. doi: 10.1080/01924036.2021.1907605. [DOI] [Google Scholar]
  52. McCann A. WalletHub; 2021. States with the fewest coronavirus restrictions.https://wallethub.com/edu/states-coronavirus-restrictions/73818 Retrieved November 5th, 2022, from. [Google Scholar]
  53. Mogavero M.C., Kennedy L.W. The social and geographic patterns of sexual offending: Is sex offender residence restriction legislation practical? Victims & Offenders. 2017;12:401–433. doi: 10.1080/15564886.2015.1084962. [DOI] [Google Scholar]
  54. Mohler G., Bertozzi A.L., Carter J., Short M.B., Sledge D., Tita G.E.…Brantingham P.J. Impact of social distancing during COVID-19 pandemic on crime in Los Angeles and Indianapolis [article] Journal of Criminal Justice. 2020;68 doi: 10.1016/j.jcrimjus.2020.101692. N.PAG-N.PAG. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Munro-Kramer M.L., Cannon L.M., Scheiman L., St Ivany A.R., Bailey J.M. Accessing healthcare services during the COVID-19 pandemic: The plight of sexual assault survivors. Journal of Forensic Nursing. 2021;17(2):93–97. doi: 10.1097/JFN.0000000000000326. [DOI] [PubMed] [Google Scholar]
  56. New York Times Five ways to monitor the coronavirus outbreak in the U.S. 2020 May 15. https://www.nytimes.com/interactive/2020/04/23/upshot/five-ways-to-monitor-coronavirus-outbreak-us.html Retrieved May 16 from.
  57. Office for Victims of Crime Rape and sexual assault. 2005. https://www.ncjrs.gov/ovc_archives/ncvrw/2005/pg5o.html
  58. Payne J.L., Morgan A., Piquero A.R. COVID-19 and social distancing measures in Queensland, Australia, are associated with short-term decreases in recorded violent crime. Journal of Experimental Criminology. 2022;18(1):89–113. doi: 10.1007/s11292-020-09441-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Piquero A., Jennings W., Jemison E., Kaukinen C.F.K. Domestic violence during the COVID-19 pandemic: Evidence from a systematic review and meta-analysis. Journal of Criminal Justice. 2021;74 doi: 10.1016/j.jcrimjus.2021.101806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Piquero A., Piquero N.L., Kurland J. A case study of domestic violence arrests during the COVID-19 pandemic in Miami-Dade County. Victims & Offenders. 2021;16(8):1077–1088. [Google Scholar]
  61. Point2Homes Downtown Detroit Demographics. 2020. https://www.point2homes.com/US/Neighborhood/MI/Detroit/Downtown-Detroit-Demographics.html Retrieved January 13 from.
  62. Rayle I. The walkability factor. 2011. https://datadrivendetroit.org/lisa-rayle-2/the-walkability-factor/
  63. Roberts N., Donovan C., Durey M. Gendered landscapes of safety: How women construct and navigate the urban landscape to avoid sexual violence. Criminology & Criminal Justice : CCJ. 2022;22(2):287–303. doi: 10.1177/1748895820963208. [DOI] [Google Scholar]
  64. Shariati A., Guerette R.T. Findings from a natural experiment on the impact of covid-19 residential quarantines on domestic violence patterns in New Orleans. Journal of Family Violence. 2022;1-12 doi: 10.1007/s10896-022-00380-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Shay K. Vox; 2020, April 10. How racism and poverty made Detroit a new coronavirus hot spot.https://www.vox.com/identities/2020/4/10/21211920/detroit-coronavirus-racism-poverty-hot-spot Retrieved May 16, 2020, from. [Google Scholar]
  66. Shen Y., Fu R., Noguchi H. COVID-19's lockdown and crime victimization: The state of emergency under the Abe administration. Asian Economic Policy Reviow. 2021;16(2):327–348. [Google Scholar]
  67. Shueh J. From debt to data: Detroit launches open data portal. 2015, March 12. https://www.govtech.com/data/From-Debt-to-Data-Detroit-Launches-Open-Data-Portal.html Retrieved May 20, 2019 from.
  68. Skogan W. In: Communities and crime. Reiss J.A., Morris N., editors. University of Chicago Press; 1986. Fear of crime and neighborhood change; pp. 203–229. [Google Scholar]
  69. Skogan W. Free Press; 1990. Disorder and decline: Crime and the spiral of decay in American neighborhoods. [Google Scholar]
  70. Stickle B. Package theft in a pandemic. 2020. https://www.ucl.ac.uk/jill-dando-institute/sites/jill-dando-institute/files/package_theft_in_a_pandemic_final_no_15_.Pdf
  71. Stickle B., Felson M. Crime rates in a pandemic: The largest criminological experiment in history [article] American Journal of Criminal Justice. 2020;45(4):525–536. doi: 10.1007/s12103-020-09546-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Taylor R.B., Gottfredson S. Environmental design, crime, and prevention: An examination of community dynamics. Crime and Justice. 1986;8:387–416. [Google Scholar]
  73. Transportation Riders United Safe(r) transit during COVID-19: Executive summary. 2020. https://www.detroittransit.org/wp-content/uploads/2020/09/Transit-During-COVID-19_-Executive-Summary-.pdf
  74. United States Census Bureau QuickFacts: Detroit city, Michigan; United States. 2022. https://www.census.gov/quickfacts/fact/table/detroitcitymichiganUS/POP010220 Retrieved November 30 from.
  75. University of Michigan School of Public Health Michigan COVID-19 Tracker. 2022. https://covidmapping.org/timeline.html Retrieved November 6 from.
  76. Wood L., Schrag R.V., Baumler E., Hairston D., Guillot-Wright S., Torres E., Temple J.R. On the front lines of the COVID-19 pandemic: Occupational experiences of the intimate partner violence and sexual assault workforce. Journal of Interpersonal Violence. 2022;37(11−12) doi: 10.1177/0886260520983304. NP9345-NP9366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Wortley R., Mazerolle L. Willan Publishing; 2008. Environmental criminology and crime analysis. [Google Scholar]
  78. WXYZ Metro Detroit becoming part of national focus as COVID-19 outbreak hot spot. 2020, April 6. https://www.wxyz.com/news/coronavirus/metro-detroit-becoming-part-of-national-focus-as-covid-19-outbreak-hot-spot Retrieved May 16 from.

Articles from Journal of Criminal Justice are provided here courtesy of Elsevier

RESOURCES