THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS The Mixed Methods Analysis of Police and Non-Police Shooters Christopher Paulgaard Bachelor of Law Enforcement Laws 4003 - Research Project Justice Institute of British Columbia Dr. Nora Houlahan Research Advisor: Daniella Sieukaran April 8th, 2022 1 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 2 Abstract The purpose of this mixed-methods research is to discover the common and differentiating factors between criminals who shoot at police or who shoot at others in Canada. This research collected secondary data from 52 participants. Two sample groups are the police shooters (shooters who target the police) and the general shooters (shooters who target nonpolice). Each sample group contained 26 participants from across Canada. Data was analyzed using a convergent, mixed methods approach. Both quantitative and qualitative factors were involved in the analysis. The research question of this study was, “How do characteristics vary for Canadian criminal shooters who target police compared to those who do not?” The characteristics are the identified past and present variables that may contribute to the criminal shooting. It was concluded that the police shooters and general shooters are similar, but they have some differences. There was a highly significant association between the police shooter and being arrested, there was a very significant association between the general shooter and the drug trade, and there was a marginally significant association between the police shooter and experiencing a traumatic event (in adulthood). The remaining variables discussed in this study were nonsignificant, but still relevant. Keywords: Criminal shooter, firearm crime, gun violence, police shooter, firearms THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 3 Table of Contents Abstract........................................................................................................................................................ 2 The Mixed Methods Analysis of Police and Non-Police Shooters .......................................................... 4 Philosophical Worldview ........................................................................................................................ 8 Theoretical Framework .......................................................................................................................... 8 Methods........................................................................................................................................................ 9 Literature Search and Review Plan .................................................................................................... 11 Data Analysis Plan ................................................................................................................................ 19 A Piori Power Analysis ...................................................................................................................... 19 Qualitative Data Analysis .................................................................................................................. 20 Descriptive Statistical Analysis .......................................................................................................... 21 Quantitative Data Analysis ................................................................................................................ 21 Results ........................................................................................................................................................ 22 Descriptive Statistical Analysis ............................................................................................................ 22 Qualitative Data Analysis ..................................................................................................................... 23 Quantitative Data Analysis .................................................................................................................. 24 Tested Assumptions ............................................................................................................................ 24 Ranked Dependent Variables ............................................................................................................ 25 Group Differences on Dependent Variables ..................................................................................... 28 Discussion .................................................................................................................................................. 29 Conclusion ................................................................................................................................................. 40 Limitations ............................................................................................................................................. 41 Recommendations ..................................................................................................................................... 42 References .................................................................................................................................................. 45 Appendix A ................................................................................................................................................. 52 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 4 The Mixed Methods Analysis of Police and Non-Police Shooters The most recent research suggests a firearm is the most common way a Canadian police officer is killed in the line of duty. Between 1961-2009, 92% of police officers killed were by firearms (Dunn, 2010). A criminal who discharges a firearm against a police officer is an immediate danger to public safety because the officer’s life, and the lives of bystanders, are put at risk. The criminal shooter who targets police also can cause harmful long-term effects on police survivors. Negative long-term effects for police who experience violence includes time loss claims and police fatalities. The fatalities are due to disorders or traumatic injuries that are related to the initial incident (Cohen & Garis, 2018). This is relevant to police survivors who have been fired upon because the incident can affect their ability to continue to work in law enforcement. The consequences that the criminal shooter creates when they target police officers is concerning. This problem needs to be researched because criminals who shoot at police harm law enforcement efforts. The criminal use of firearms against the police is an important issue. This study will be comparing two types of samples. The cases in which firearms are used against police represent the police shooter sample (police shooters), and cases in which firearms are used against people who are not police represent the general shooter sample (nonpolice shooters). The general shooter sample poses an additional problem to society. Violent firearm offences, including the discharge of a firearm with intent, have increased annually over the last six years in Canada (Moreau, 2021). The criminal discharge of a firearm is concerning because firearms can be used to aid criminals in their homicides, robberies, sexual assaults, and assaults against the public (Moreau, 2021). The contribution that the criminal use of firearms makes to homicides is of particular concern. Firearms have been the most common method of Canadian homicide annually from 2016-2020 (Statistics Canada, 2021a; Statistics THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 5 Canada, 2016). Prior to these time frames, firearms have been the most common or second most common method of homicide since 1987. This is because rates of homicide by stabbings compete with firearm homicides for the place of being the most common homicide method (Beattie et al., 2018). In addition to the prior issues that the criminal use of firearms has caused, there are also long-term effects on the crime survivors. Long-term harm of firearm-related crimes on Canadian victims may include mental health issues, substance use issues, reduced economic productivity, and health care costs (Zhang & Qin, 2012). The criminal use of firearms against the public and the police is a Canadian security concern. It has violent effects, and it negatively impacts the victims. The recent change of the increase in criminals discharging firearms have resulted in the need for a comprehensive research approach that may help improve future public safety strategies. To date, there is very limited to no research that compares a police shooter sample (PSS) to a general shooter sample (GSS). This is because there has been minimal research about this topic, and the research that does exist has minimal relevance. A literature review first proved there is minimal research about this topic. Of the 151 potential results reviewed, 148 of the articles were not relevant to the current study and only three had minimal relevance. There is a lack of research comparing the GSS and PSS because of the lack of relevant articles to this subject field. Of the relevant articles, many were very limited in their relevance to the context of comparing the PSS to the GSS. The first article discussed police shooters in the context of the United States (Issitt, 2021). Issitt (2021), discussed two incidents of police shooters, and the factors about these shooters were identified. This research was very limited because it is only about two case studies of police shooters from the United States, they are about politicallyrelated shootings (so not necessarily representing the typical police shooter), and no comparison THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 6 to a shooter who targeted nonpolice was done. This does not contribute to the understanding of the comparison of Canadian police shooters and Canadian general shooters. The next article is by Logan (2021), and it was in the Canadian context. This article was about a general overview of factors that predisposed someone to violence with firearms. This had some relevance to shooter profiles because it analyzed their risk factors. However, it offered very little to contribute to the current study. No comparison of the profile of the police shooter and general shooter was made, and no discussion of police shooter was conducted. The last article reviewed did mention some police shooter cases in the Canadian context. The final article reviewed was by Business Vancouver (2016) (that sources a 2015 Prince George Citizen article). It had some relevance because the profiles of Canadian police shooters were discussed. This article also has major limitations to how it contributes to the current study. It appears to be an opinion piece, and it does not compare police shooters to general shooters. All three articles established risk factors about the GSS and PSS. Overall, the risk factors of police shooters included having a criminal record, having multiple firearms bans issued by the court, access to illegal firearms, having a personal attitude of being wiling to use a firearm against a police officer (Business Vancouver, 2016), access to legal firearms, being a military veteran, the shooter being personally upset about the killings of African-Americans by police officers, and the shooter having a racist attitude towards White police officers (Issitt, 2021). In a more general overview, risk factors of shooters included having an untreated severe psychiatric disease, substance abuse, a history of violence, committing intimate partner violence, having a criminal record, having suicidal behaviour, having mental health issues, risk factors in the school environment, family environment, in the individual, and in community factors (Logan, 2021). Each of these articles had some relevance to the current study, but it was very limited. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 7 The knowledge about the comparison of police shooters and general shooters is very limited, or not publicly in existence, based on the current literature review. The author of the current study is not aware of any study that has compared these two groups in the Canadian, or other contexts. This represents a major gap in the current academic knowledge of comparing police shooters to general shooters. However, the current study has made it possible to address these research gaps. The purpose of this mixed-methods research is to discover the common and differentiating factors between criminals who shoot at police or who shoot at others in Canada. This research will collect secondary data and analyze it using a convergent, mixed methods approach. Both quantitative and qualitative factors will be involved in the analysis. This study will aim to achieve the purpose by studying 26 individuals from each group. A sample size of 26 individuals for each group was decided upon because a priori power analysis determined this was the sample size required to detect large differences between the groups. The factors of each shooter will be identified, then the characteristic averages of each sample group will be discovered, and then the sample groups will be compared. The comparison will reveal if there is any difference between the two samples. The appropriate research question to address this area of research is, “How do characteristics vary for Canadian criminal shooters who target police compared to those who do not?” The characteristics are the identified past and present variables that may contribute to the criminal shooting. A criminal shooter is a person who is convicted criminally of discharging a firearm at another (either a police officer or nonpolice officer). The researcher hypothesizes that criminals who target the police will have a greater chance of having a prior criminal record than those who target nonpolice persons. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 8 Philosophical Worldview The research was conducted with a postpositivist world view. This means that the researcher believes the data results can be tested and reviewed to make accurate conclusions. This worldview believes objective results can be produced with no major subjective interference (Creswell & Creswell, 2018). This is because statistical tests and educated-based reasoning will cause the effects of subjectivity to have such minimal effect that it will not cause a change to the overall conclusion. That is how the postpositivist view was applied in this research. An example this perspective can be given. This view recognizes that research (especially quantitative results) can explain things to make accurate conclusions. For instance, the reported number of shots can be measured to see if one group has a higher average of shots fired than the other group. However, it is acknowledged that results can never be 100% certain due to the limitations of understanding every existing variable, but it is still fair to have a high degree of certainty in objective results. Theoretical Framework The theoretical framework of this study includes the use of an inductive strategy, and incorporating a social science theory (Creswell & Creswell, 2018). This study will use an inductive strategy because the research strategy is designed to gather data (variables), look for patterns (such as frequencies), and a theory will be developed about the two samples (Blackstone, 2012). Additionally, the inductive strategy was used because a deductive approach would not be effective in this research context. A deductive approach is not ideal for this study because a deductive approach tests a hypothesis created on previous theories of the studied group (Blackstone, 2012). A hypothesis could not be generated as no previous theories about the populations was found in the limited prior research of this study field. There are many social THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 9 science theories that helped inform this research. These include sociological views, biological views, biosocial views, psychological theories, the moral development theory, and the social learning theory. Some examples of these theories will be given, all examples are in the context of how the factor affect crime. Sociological views focus on factors such as employment, school attachment, economic status, child maltreatment, and parental behaviour (such as drug use). Biological views focus on brain chemical interactions (such as serotonin and testosterone). Biosocial interaction views how social interactions can affect our biology, and psychological theories focus on the cognitive function (Schmalleger & Volk, 2018). Even though these social science views were considered, they were not the target variable type of collection. Any variables relevant to the offender and the offence were collected in order to minimalize a bias focus when identifying variables. Methods Research Design and Rational Secondary data was collected from the Canadian Legal Information Institute (CanLII) database using the CanLII search engine (CanLII, n.d.). Using the CanLII database for research has many strengths. First, the cases of criminal shooters are easily accessible because they are published publicly, and they can be accessed through the internet. This is good for research because the cases can be collected and analyzed in a timely manner. The content of the cases in CanLII is another strength. The relevant factors of the convicted person are normally discussed by the judges and lawyers. Having the relevant factors of the convicted person makes the collection of relevant data to this study achievable because this study aims to identify variables about the criminal shooter. Furthermore, the CanLII database does not have weaknesses that an alternative research data source, such as news reporting database may have. This is because THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 10 CanLII does not publish their content in a biased manner when compared to a news database. CanLII does not choose what kinds of criminal offences they will report on because any offence will go to court, as long as the police submit the charges to crown, and crown approves of the charges. A news agency has greater discretion on what they will create reports on. A United Kingdom study found that news agency discretion can cause a bias in their database because relevant cases may not be included if they are not considered to be sensational (Gekoski, et al., 2012). CanLII was chosen because it is easy to collect data from, and the data is not published in a way that is bias like a new reporting database. Defining the terms used in this study is important to understand before the discussion of the search begins. It is important to operationalize the terms criminal, shooting, and police officer to demonstrate the research analyzes what it claims to study. A criminal is any individual found legally guilty (also known as convicted) of a crime in Canada. A shooting is defined as the discharging (firing) of a firearm (a rifle, pistol, shotgun, or other type of firearm). This study does not define the discharging of a pellet rifle or a trades tool (such as a nail gun) as a firearm shooting. A police officer is a municipal, regional, provincial, or federal peace officer who has the duty of maintaining peace and law in the public. Provincial sheriffs, conservation officers, correction officers, and bylaw officers are not included in the definition of a police officer. In reference to the research question, criminal shooters who target people other than police, are known as “Others.” “Others,” is defined as any individual other than a police officer. The search terms used is another important part of the search methodology. A systematic approach was used for data collection. First, the case would be collected by using the key terms in the CanLII search (later discussed), results were sorted from newest to oldest, results were sorted to attempt to reflect to population distribution of Canada as shown in THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 11 table 1A and table 2A (Appendix A). Each case was reviewed beginning with the first result, an inclusion and exclusion criteria was applied, and the review for cases would end when the population for that jurisdiction was met, or if the results were all reviewed. A “Coding Criteria Method” (later discussed) was applied to each case to detect the status of each variable. Literature Search and Review Plan The search terms (key words) used for each group differed slightly, but each represented the type of shooter concept accurately. The general criminal shooter sample key words included: "firearm" or "gun" or "handgun" or "pistol" or "rifle" or "shotgun" and "discharg*" or "shot" or "fired at" and "convict*" or "guilty" and "intent to cause bodily harm" or "reckless endangerment" or "intent to wound." The key words for the police shooter sample included: "shot at police" or "shooting at police" or "shot police" or "shot detective" or "shot at corporal" or "shot corporal" or "shot at constable" or "shot constable" or "shot at officer" or "shot officer" or "fired at officer" or "firing at police" or "firing at corporal" or "endangering police" or "discharg* at police" and firearm or "gun" or "handgun" or "pistol" or "rifle" or "shotgun" and "convict*" or "guilty." These search terms exclude some police ranks (such as sergeant). Other ranks were attempted to be used in the search terms, but the CanLii search engine would result in fewer, less relevant results. This may be due to some search engine issue. The search terms used for the GSS and PSS are slightly different, and there is reason for this. The search term lists used for both samples are slightly different from one another. It was attempted to have the same search terms (other than the ones relevant to the police). However, this caused data result issues. The results were either too broad (14,924 results for the general criminal shooters) or too narrow (21 results for the police shooter sample). These search terms ultimately had a better case results number that allow for reliable case gathering counts, while THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 12 reducing an overload of thousands of results, as shown in table 1B (Appendix B). These search terms represent the same overall concepts, and are the appropriate for the shooting contexts. All of the key words represent the factual or legal phrases frequently involved in criminal and police shootings. There was an additional search strategy used for the Police Shooter Population because the first search strategy did not result in the required number of samples. Another search strategy for the PSS was used because the original search strategy did not result in enough samples that reflected Canadian demographics. This additionally PSS search strategy was more broad search term which was, “shot the firearm at police” (no quotations used in search term). This gave results in regions to collect samples from that were priorly lacking samples. The results were not as relevant because this was a broad search term. Due to time, the researcher could open each case and read the details. Instead, the researcher did a research style that is somewhat like conducting an abstract review. After searching on CanLII, a long list of results appears. Each result is numbered with the case name. Under the case name is a brief overview of the areas of the case that are considered relevant to the search terms used (referred to as the “relevance section” by the researcher). The relevance section acted as the summary view of the case. Each relevance section was scanned to see if any obvious police shooting case existed. If it appeared to be a PSS case, the case of interest was further viewed, and the inclusion and exclusion criteria was applied. The additional search strategy for the PSS helped the researcher collect the required number of samples, while still maintaining an effective research strategy. One last search strategy was required to complete the PSS. There was difficulty finding a PSS case for Nova Scotia, but this issue was resolved. The researcher found previous case law of a police shooter case when reviewing a case from the “shot the firearm at police” search. This case law was used and the process of how the case was THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 13 included is documented in the inclusion and exclusion spreadsheet. Once cases were searched for, the inclusion and exclusion criteria became an important aspect of the research. The inclusion criteria of the criminal cases includes: The person(s) being convicted (guilty) in court of criminally discharging a firearm (in relation to a circumstance where the criminal shoots at another person), or having the intent to harm another person with a firearm (in relation to discharging the firearm), or recklessly discharging a firearm (in relation to a circumstance where a person is shot at), or the firearm is used and discharged in attempting to unlawfully cause bodily harm (against another person), or using a firearm to cause reckless endangerment (in relation to discharging a firearm at a person), or discharging a firearm with intent to endanger life (of another person), or aggravated assault (by firearm discharge against another person), or attempted murder (by firearm discharge), or second degree murder (by firearm discharge), or manslaughter (by firearm discharge), or in a case if the person is found guilty of careless use of a firearm (in relation to them shooting the firearm at someone else). There are many legal words in this study for how a person violently uses a firearm against another. This is because judges have multiple ways to apply the law in finding a person criminally responsible of discharging a firearm against someone else. Furthermore, in some cases, if the criminal discharges the firearm at a home or place where they believe or knew a person went into, that may be included in this study (this will be further justified why it is included or excluded on a case-by-case basis). The legal offence for this is known as being reckless as to whether another person was present in the place and discharging the firearm into that place. Additionally, the police shooter sample will have to meet the criteria of a police officer being shot at (previously mentioned). THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 14 Legal cases will be excluded if they meet one or more of the following criteria: When the shooting event occurred during the year 2000 or in prior years, the person charged is acquitted (innocent, not guilty, not convicted), if the person is acquitted later in an appeal case of all criminal charges related to the act of discharging the firearm against another person, when the person shooting is not shooting at any other person or in the direction where other persons are known to be, when the shooting charge is due to a hunting accident, or if the person was legally found to have acted in self-defence. The age of the people in the samples, location of the offence, and the time frame of the incident is another important part of the inclusion and exclusion criteria. This research has considered the relevance of the offender’s age, offence location, and time frame. This study will analyze offenders of any age (young offenders and adult offenders) because anyone who shoots a firearm at another person is a risk to public safety. Shootings that occur anywhere within Canada will be included because this study has the goal of focusing on Canadian shooters. Shooting incidents that occur outside of Canada will not be included. The CanLII search sorted cases from newest to oldest because the ideal timeframe (but not exclusive) of collection will be from 2011-2022. This is an important period to study because there was a change in firearm homicide trends. During this time, shooting homicides in 2011 and 2013 were the lowest in almost fifty years (Statistics Canada, 2012; Statistics Canada, 2014). However, the situation changed, and there has been an overall increase in firearm homicides. From 2016-2020 there have been over 250 firearm homicides each year. In 2020, there was over double the amount of shooting homicides than in 2013 (Moreau, 2021). Conducting research on this particular time frame will make information available for future researchers who may want to compare the criminal shooters of 2011 to 2020. However, cases were collected before this THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 15 timeframe in order to meet the sample size requirement of this study. For example, Newfoundland and Labrador’s most recent reported police shooting case on CanLII was in 2010. 40 of the case incidents were from 2011-2020, while 12 of the shooting incidents were from 2004-2010. Including the cases that occurred prior to 2011 is necessary because it improved the case collection for areas where there otherwise would have been no representation. There is also is two types of exemptions to the inclusion criteria in regards to the shooter being criminally charged. In R v Hennessey (2009), the shooter was killed during a shootout with police, so the shooter could not be criminally trialed. Additionally, in R v Morrill (2016), the shooter was found not to be criminally responsible due to mental disorder. Both of these cases will still be included because both individuals shot a firearm at police, and they shot at the police without any legal justification. These two people still represent a threat to law enforcement as they endangered the lives of police officers by shooting at them, so they will still be included. The cases were gathered from the newest to oldest (by a date sorting function on CanLII) cases. This was done because the research aimed to analyze a more recent time period (20112022, as previously discussed). The CanLII sorting of province or territory was also used. This geographic sorting allowed the cases to be collected in a manner that helped better reflect the distribution of the Canadian population as demonstrated by table 1A and table 2A (Appendix A). The cases found by this search strategy all underwent the inclusion criteria review (mentioned previously in the search methodology). If this case has a “history” link that refers to the “sentencing” case, the sentencing case will be where the data is gathered from. The sentencing case document is the most reliable document to gather data from because they typically contain the most relevant data to the variables involved. The offenders’ circumstances, and the circumstances of the offences are frequently discussed in sentencing cases. In comparison, a THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 16 legal case that discusses the guilt or not of the accused may not fully account for the accused’s circumstances until they are sentenced. However, cases other than a sentencing case may be used if the sentencing case is not available, or if the other types of cases contain important variable information that the sentence case does not (such as the type of firearm used in the offence). The legal cases analyzed will also reference prior legal cases (case law) that are similar. The case law references may be used to create a “reserve list” if the necessary amounts of cases cannot be collected through the search terms. A reserve list case will undergo a full-text review if it is included. If this study does use a reserve list, a bias limiting approach will be explained and used. The researcher contacted his research supervisor when the shooting case was difficult to initially decern for including or excluding. Cases that were uncertain for being included was a rare event, but it did happen. In the R v Kakfwi (2018) case, the shooter was in a building. The shooter shot a warning shot into the air when he heard someone (a police officer) outside the building. The judge did not find that he meant to shot at the officer, and the discharged round did not go near the officer. The research supervisor assisted in reviewing the exclusion of this case. Having the research supervisor as a secondary review helped enhance the internal reliability of the case inclusion. The inclusion and exclusion process were conducted for both samples. When the initial search was extinguished of results, a broader search term was used. The broader search term resulted in many results, so a scan of the key word area of each case in the result list was conducted. The police shooter sample had a total of 134 cases reviewed (Nunavut 1, North West Territories 22, Yukon 19, Newfound Land and Labrador 2, Prince Edward Island 5, Nova Scotia 5, New Brunswick 6, Quebec 5, Ontario 35, Manitoba 1, Saskatchewan 6, Alberta 8, British Columbia 19) and 1,227 case key word areas of scanned (Nova Scotia 129, New Brunswick 31, THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 17 Quebec 126, Ontario 500, Saskatchewan 14, Alberta 427). General shooter sample had a total of 56 cases reviewed (Nunavut 1, New West Territories 3, Yukon 2, Newfound land and Labrador 2, Prince Edward Island 1, Nova Scotia 1, New Brunswick 1, Quebec 9, Ontario 26, Manitoba 2, Saskatchewan 1, Alberta 4, British Columbia 3). A total of 190 legal case documents were reviewed and 1,227 case key word areas of legal case documents were scanned to generate the 52 required samples (participants) for this study. The studied samples generated many qualitative results. Once the case met the inclusion criteria, it was downloaded as a PDF (portable document format) file. The researcher then read through the PDF file and applied the coding criteria method as shown in Appendix C. The Coding Criteria Method is a set standard of how the researcher reviewed each case. This ensured that there was a consistent way to review for each factor in all the cases. The factors (variables) were discovered, and then added to the Coding Criteria Method, then the researcher would go back to prior cases and scan for that same variable. If the variable was not mentioned, certain requirements were needed in order for it to be labeled as “no” (not present) or as a 99 (missing value). Reference Appendix C for a comprehensive overview of how each variable was reviewed. The data selection assessment (the coding criteria method) worked well because it allowed the research to apply a consistent way of scanning and coding data. Its development may not be as refined as the researcher would have preferred due to time constraints. However, it still completes the overall purpose of deeming if a variable is present, not present, if the value is missing, or if the value is not applicable. The main challenges of this research were acquiring a sufficient number of cases, and budgeting enough time to conduct the discussion. It was initially difficult to find a sufficient number of cases because many combinations of the appropriate search terms had a different THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 18 number of results. A balance of the appropriate search terms and having enough results was a major task. Additionally, acquiring a sufficient number of police shooter cases was more difficult to find than general shooter cases. Using additional databases other than CanLII may be of future interest to address this challenge. Ethical Issues The researcher was first concerned if any external ethical approval was required for this study. The researcher’s supervisor verified with the JIBC (Justice Institute of British Columbia) Chair of Research Ethics Board and the Program Manager of the Law Enforcement Studies Programs to confirm that ethical approval was not required for this study. Ethical approval was not required because this study collects secondary data that is publicly available. However, there are still ethical concerns with public data. The ethical concerns included the mislabelling of an accused person as a criminal and of researcher bias. The mislabelling of an accused person as a criminal is a concern. The author addressed this by assuring that all of the individuals labelled as a criminal are convicted of their criminal act, without any successful appeal. Appeals will be reviewed in the history tab of each CanLII case. This made sure that the study correctly labeled who was a criminal. Researcher bias is another issue that was effectively addressed to maintain the credibility of the findings. The researcher has a firearms licence and tends to view articles that are more bias towards maintaining a firearm licencing system. Additionally, the researcher will disclose that he is a member of a private Facebook group called “Canadian Coalition for Firearm Rights.” Some may view this as a risk because they may believe the researcher will avoid including data that discredits licenced firearm owners. This issue was addressed because having a firearms licence is not part of the inclusion or exclusion data, and the included cases will have their overall data reviewed, regardless if a firearms licence status was noticed during THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 19 the initial inclusion and exclusion stage. Furthermore, the coding criteria (Appendix C) sheet has been designed to safeguard from any over estimation of criminals who do not have their firearms licence. The firearm licence factors had to be mentioned more directly than many other factors in order for it to be concluded as existing or not existing in each case. If this higher requirement was not met, then it would be coded as a missing value. Lastly, the author would like to mention that he did not grow up in a family that favoured firearms, and he has reviewed discussions over many years that are both for and against civilian firearm ownership. Though the author may have personal views about firearms, that did not become an ethical issue during the collection of the relevant data. Lastly, the author has also reviewed concepts that raise awareness of individual bias, and he will ensure that bias does not cause any change to the future results or to the discussion section (Pannucci & Wilkins, 2010). Furthermore, the researcher has prepared for any other potential ethical issues by reviewing relevant research policies (3401-3405, and 3407) (Justice Institute of British Columbia, n.d.). These are the major ethical issues that can have all been effectively addressed. Data Analysis Plan A Piori Power Analysis For independent samples two-tailed t-test analyses for this study’s scale data, and a piori power analysis using Cohen’s (1988) convention of 0.8 for a large effect size (r) was conducted using G*Power 3.1. In order to sufficiently control for type II error with an alpha level (i.e., type I error) set at 0.05 and with a desired statistical power level of 0.8, the minimum sample size needed to detect a large effect was determined to be 52. This sample size is reflected by the 26 cases of the first sample (police shooters) and 24 cases of the second sample (general shooters). One of the 24 cases of the GSS had three participants in it (R v Baugh, 2021), so that is why only THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 20 24 cases were required. In the event that data was non-normal, a power analysis was also conducted for the nonparametric equivalent of a t-test – the Mann-Whitney test. Using the same parameters as outlined above (such as the power level of 0.8), a minimum sample size for a logistic distribution was calculated to be 48 and 36 for a Laplace distribution. The planned sample size of 52 exceeded the minimum sample size for both of these non-normal distributions. Qualitative Data Analysis The common themes (variables) were identified by reading through each case document (PDF), highlighting the variable (yellow highlight), and then determining if it already fell into an identified variable category, or if a new variable was needing to be generated (detail of what variable each yellow highlight is can be found in the PDF comment). An example of the coding process can be given from R v Cornell (2014). “Cornell’s parents separated when he was very young [childhood divorce variable] … Because of his mother’s drinking, Mr. Cornell lived with his paternal grandparents between the ages of three and five [childhood inadequate caregiver variable]”. Each variable has a definition to whether a particular fact meets its inclusion criteria. This allowed the variables to be detected and counted for the quantitative process. The variables were grouped into categories by the researcher to improve the review process of the variables. The grouping of the variables was based on their relevance to one another. For example, demographic factors included gender, age, ethnicity, among others. Childhood factors contained childhood trauma, childhood inadequate caregiver, among others, as shown in table 1D (Appendix D). The raw frequency (how much of a sample had the variable present) was calculated by the number of participants who had the factor, divided by the number of applicable responses, and multiplied by 100. For instance, 5 (people who had the factor present) divided by 25 (of the total applicable number of people with that specific factor) multiplied by 100 gives a THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 21 frequency of 25%. The frequency of every variable was calculated to determine the top five and bottom five frequencies for the two samples as shown on table 1E and table 2E (Appendix E). This specific research focused on the frequency rates. However, some other researchers may give more value to the reporting of the raw numbers (such as 5/10 participants and 20/25 participants, rather than a 50% and 80% frequency rate). For raw numbers, see table tables 1F-6F (Appendix D) The raw number charts are reported in this study because readers of this research may also value what that data states. Descriptive Statistical Analysis A descriptive statistical analysis was conducted to determine the percentages for nominal data (e.g., marital status), and means for scale data (e.g., age). Additionally, where applicable, standard deviations and minimum and maximum values were conducted (e.g., age). This process was applied to the following demographic variables: Gender, ethnicity, education, age, and marital status. This was done in order to reduce the chances of making a Type I error and reduce the potential for false positives, relevant tests were conducted to determine if there were any significant differences between the two samples on the aforementioned demographic variables. Quantitative Data Analysis All data was cleaned prior to running any quantitative analyses. For example, the data sheets were manually reviewed for any blank cells, gaps, or if any numbers were incorrectly inputted (such as a number that would not be properly coded). In addition, the assumptions of (1) independence of data, (2) normally distributed scale data, and (3) minimum cell sizes for nominal data were checked to ensure that there were no violations. Assumptions were checked for each study variable and for each of the two samples, where appropriate (the two sample groups were analyzed together and separately). First, the assumption of independence of data THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 22 was considered during data collection. Second, the assumption of normally distributed scale data was tested by visually inspecting normal Q-Q plots and histograms with normal curves. Datapoints clustered around the line of normality (Q-Q plots) or the normal distribution curve (histograms) support the assumption of normal data. Third, the assumption of minimum cell sizes for chi-square analyses of nominal data was checked. The standard convention for minimum cell sizes is greater than five per cell (Field, 2009). Group differences were tested for all study variables. It was planned to run independent samples two-tailed t-test analyses for scale data, and Mann-Whitney tests if the data is nonnormal. Chi-square analyses were used for nominal data, and odds ratios were calculated for significant results. For all analyses, missing data was excluded pairwise, meaning a case was only excluded from calculations when there was a missing value for the variable being tested. Results Descriptive Statistical Analysis This study had a total of 52 participants from 50 different legal cases. 98.08% of the participants were male (n= 51) and 1.92% of the participants were female (n= 1; participant in the PSS sample). Participants had an age (in years) range of 15 to 57, and a mean age (average) of 27.36 (SD= 9.92). The ethnic representations of the studied group included (of those who reported in, n= 18) Caucasian 11.1%, Indigenous 61.1%, Middle Eastern 5.6%, South Asian 5.6%, Mixed 5.6%, and Other 11.1%. The education reporting (n= 26) showed 23.1% had less than high school, 38.5% had some high school, 23.1% completed high school, 3.8% had some college or university, 7.7% completed college, and 3.8% completed an undergrad university degree. The marital status reporting (n= 32) displayed that 40.6% were single, 46.9% were in a common law relationship, and 12.5% were married. Lastly, every province and territory in THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 23 Canada has some level of representation. Overall, the two groups were overall, similar in their gender, education, martial status, age, ethnicity as shown in table 1G (Appendix G). Qualitative Data Analysis Data coding was conducted in all of the cases to reveal qualitative themes. The themes were then put into groups. The grouped variables are classified together because they are part of a similar category. There were ten categories generated from the common themes. The 89 variables generated from the research are classified in these categories, as shown in table 1D (Appendix D). The type of the criminal offence of the shooting has been analyzed, rather than the section number. The section numbers will not be analyzed because what the section numbers represent may change over time. The cases collected are from 2004-2020, and there have been changes to the Criminal Code during that time period (Government of Canada, n.d.). Therefore, reporting the Criminal Code section number would not provide meaningful or accurate information. The written section of the charges was mentioned in many of the legal cases. Of the 52 participants, there are 15 ways that a criminal shooter is convicted. The types of convictions that a shooter can be convicted of in relation to the shooting are shown on table 1H (Appendix H). There are six incident categories that a police officer was responding to before being shot at. These incident types include • Violence related (42.31% of incident response types, n= 11) • Other (19.23% of incident response types, n= 5) • Relationship related (11.54% of incident response types, n= 3) • Theft related (11.54% of incident response types, n= 3) THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS • Drug related (7.69% of incident response types, n= 2) • Suicide related (7.69% of incident response types, n= 2) 24 Other includes, person of interest to the police, a call for assistance, police warrant, person endanger, and a male walking in public carrying rifles and wearing camouflage. There are four categories that can be related to why to nonpolice officer was targeted • Drug related (40.91% of incidences, n= 9) • Social dispute (31.82% of incidences, n= 7) • Violence related (22.73% of incidences, n= 5) • Alcohol related (4.55% of incidences, n= 1) For context, violence related can be a violent conflict between individuals, or a violent event. In both instances, the situation is violent related before the firearm is criminally discharged. An example of violence related is the R v P.B. (2021), in which the criminal discharged a firearm during a robbery (a robbery is theft with violence). Quantitative Data Analysis Tested Assumptions Prior to running any analyses, the assumptions of (1) independence of data, (2) normally distributed scale data, and (3) minimum cell sizes for nominal data were checked to ensure that there were no violations. Assumptions were checked for each study variable and for each of the two samples, where appropriate. First, the assumption of independence of data was accepted to be true because sound data collection strategies were employed (i.e., data were independent as each case was assigned to only one of the two samples). Second, the assumption of normally distributed scale data was tested. In both samples, normal Q-Q plots show that some data points THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 25 deviated slightly from the line of normality, and histograms revealed some non-normal distribution curves. Therefore, it was decided to use Mann-Whitney tests, the nonparametric equivalent to t-tests, for scale data. Two outliers for age (both aged closer to 60) were detected, but they were included (R v Crockwell, 2013; R v Fobister, 2009). The nonparametric test calculations adjusted for these data abnormalities. In order to prevent type II errors due to small sample sizes, marginally significant results were accepted (p < 0.08). Third, the assumption of minimum cell sizes for chi-square analyses of nominal data was checked. For any analyses where the cell sizes were fewer than five, Fisher’s exact tests were utilized. Ranked Dependent Variables The five highest dependent variable frequencies (non-significant data except for intergenerational trauma) for the PSS are (the list exceeds five dependent variables as there were variables with the same frequency precents. Data from table 1E, Appendix E): • Violent history (66.7%) • Drug use history (61.1%) • Criminal record (61.1%) • Intergenerational trauma (57.1%) (Indigenous specific variable) • Unemployment current 57.1% • Unemployment history (56.3%) • Alcohol use history (55.6%) The five highest dependent variable frequencies (non-significant data) for the GSS are (the list exceeds five dependent variables as there were variables with the same frequency precents, data is table 2E, Appendix E): THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS • Childhood trauma (68.8%) • Single parent (68.8%) (childhood specific variable) • Father relation issue (66.7%) • Criminal record (64.7%) • Violent history (61.1%) • Unemployment history (58.3%) 26 Refer to table 3E (Appendix E) for the highest frequency among both groups combines. Shooting Incident Details Among the PSS, the police shooter shot first 96% of the time, while the police shot first 4% of the time. Among the GSS the general shooter shot first 100% of the time. Among the PSS, the police had a 61.54% chance of discharging their firearm during the incident as shown in table1I (Appendix I). The total amount of shots fired by each group was calculated, and their average shots fired per incident was documented as shown in table 2I (Appendix I): • General Shooters total shots fired: 74 (an average 3.36 shots fired per incident) • Police Shooters: 45 (an average of 2.37 shots fired per incident) • Police Officers: 13 (an average of 2.6 shots fired per incident) Shooting accuracy means the number of shots that hit the target when the shooter aimed at the target. Table 1I (Appendix I) displays the shooting accuracy of each group was: • Police: 46.15% • Police shooter: 26.67% THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS • 27 General shooter: 19.5% Each groups rate of survival, or rate of experiencing at least one or more deaths (in the context of police incidents), is displayed in table 1I (Appendix I) and is below: • General shooter survival rate 100%, death rate 0% • Police shooter survival rate 92.31%, death rate 7.69% • Nonpolice target survival rate 80%, death rate 20% • Police survival rate 84.62%, death rate 15.38% (The 15.38% is broken down as follows: 3.8% chance of one police death, 3.8% chance of two police deaths, 3.8% chance of three police deaths, and a 3.8% chance of four police deaths) The types of firearms used by the shooter in both the PSS and GSS was documented and is shown in Figure 1. Figure 1 Firearms Used by PSS and GSS THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 2% 28 Pistol Shotgun 17% 2% Bolt Action Rifle 2% 39% Semi-automatic Rifle Rifle- Unknown Type 13% Pump Action Rifle Automatic Rifle 8% 15% Homemade Firearm Unknown 2% Note. This chart represents the types of firearms used by both the PSS and the GSS (n=52). This is rounded to the nearest percent. When police are shot at, it typically is by a shooter with a long-gun (a rifle or shotgun), and not a pistol. 61.90% of police shooters use a long-gun, and 38.09% of police shooters used a handgun. Police are 16.45% more likely than the nonpolice to be targeted by someone using a rifle or shotgun. Criminals who target nonpolice use pistols 54.54% of the time, and a rifle or shotgun 45.45% of the time, as shown on table 3I (Appendix I). The reporting participants (n= 29; PSS= 14, GSS= 15) demonstrated that no shooter had a firearms licence, as shown on table 1J (Appendix J). Group Differences on Dependent Variables The results of the Mann-Whitney analyses testing group differences on scale dependent variables are shown in Table 1K (Appendix K). Three out of ten tests were run because some factors were mutually exclusive to one group, so a test comparing both groups could not be done for those variables. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 29 The results of the chi-square analyses testing group differences on nominal dependent variables are shown in table 1L (Appendix L). Contingency tables for all significant chi-square analyses can be found in Table 1M-3M (Appendix M). There was a highly significant association between the police shooter and being arrested, x2 (7) = 44.98, p<0.001****. Based on the odds ratio, the odds of police shooters were 6.75x times higher to be arrested than general shooters. There was a very significant association between the general shooter and the drug trade, x2 (1) = 8.6, p<0.006***. Based on the odds ratio, the odds of general shooters were 3.56x times higher to have the drug trade be relative to the shooting incident when compared to the police shooter. There was a marginally significant association between the police shooter and experiencing a traumatic event (in adulthood), x2 (1) = 4.5, p<0.075*. Based on the odds ratio, the odds of a police shooters were 2.94x times higher to experience a traumatic event in adulthood prior to the shooting incident than the general shooter. Discussion A large amount of data has been discovered that can help answer the research question, “How do characteristics vary for Canadian criminal shooters who target police compared to those who do not?” Overall, the PSS and GSS are similar, but they have some differences. First there were two statistically significant differences (not including arrest as the arrest in not a characteristic). The GSS had a very significant association with the drug trade being involved in the incident, and they were 3.56x times more likely than the PSS to have the drug trade involved in their shooting incident. The GSS appears to be more involved in the drug trade than the PSS. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 30 Next, there was a marginally significant association between the police shooter and experiencing a traumatic event in adulthood. Police shooters were 2.94x times more likely to experience a traumatic event in adulthood prior to the shooting incident than the general shooter. The PSS appears to be more affected by adulthood trauma than the GSS. The non-significant frequencies also displayed some differences, but many similarities. There were some similarities and differences for the top frequencies (non-significant data) of each group. Both the PSS and GSS had a violent history, criminal record, and a history of unemployment in their top frequencies (table 1E, table 2E; Appendix E). The PSS had some factors in their top frequencies that the GSS did not have. Factors in the top frequencies of the PSS included drug use history, intergenerational trauma, unemployment current, and a history of alcohol use. These factors were not in the top frequencies of the GSS. Factors in the top frequencies of the GSS included childhood trauma, a single parent in their childhood, and father relation issues. Even though some of the top frequencies were similar, the PSS is more affected by intergenerational trauma, unemployment current, a drug use history, and an alcohol use history. In comparison, the GSS is more affected by childhood trauma, having a single parent at some point in their childhood, and by father relation issues. A total of 60 variables from table 1J (Appendix J) that all relate to the profile of the offender (not including their demographics, and other factors such as shots fired). Both the PSS and GSS had eight of the exact same frequencies (non-significant). These same frequencies include • Childhood school trouble, both 46.7% • Mother relation issue, both 33.3% • Foster care, both 10% THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS • 31 Firearm licence, dependent trouble current, dependent trouble history, extended family trouble current, extended family violence history, all 0% Frequencies (non-significant) unique to the PSS include (5 factors) • Aggressive music obsession • Extended family violence • Firearm obsession • Sleep deprivation • Violent videogame addiction Frequencies (non-significant) unique to the GSS include (6 factors) • Associate violent history • Community partner violence • Family violence history • Nonfamily trouble history • Misogyny • Racism There were 41 remaining frequencies (all non-significant, except for traumatic event [adulthood] and drug trade current) were shared at varying frequencies between the PSS and the GSS. Of the 60 variables (table 1J; Appendix J), it can be concluded that both the PSS and the GSS are similar, and they have some differences. They share 68.33% (41/60 variables) of the same variables at different frequency rates, they are 13.33% exactly the same (8/60), and they are exclusive in 18.33% (11/60) of the remaining variables. These percentages are subject to some change if the demographic and circumstance of incident (such as shots fired) were to be THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 32 included. It is important to discuss the variables in greater detail as to why they may be relevant to each population and their shooting action. Statistically Significant Variables Discussion The highly significant association of the police shooters and being arrested occurred because police are able to respond to the shooter instantly. Police can respond immediately to the police shooter because a police officer is the target, and that officer is in the immediate area ready to engage with and arrest the shooter. Furthermore, police backup to assist the officer under fire would bring more officers to the scene. Having an increased number of officers responding also increases the chance of arrest. If the criminal managed to temporarily evade police, the police would pursue the shooter and other police units in the region would be alerted of the shooter. The immediate response of the police under fire, having police back up called, using police pursuit tactics, and alerting police in the region about the shooter all contributed to the higher arrest rate of police shooters. In comparison, general shooters who targeted nonpolice have a greater chance of not being arrested during the initial shooting incident. The main reason for this is that police were far less likely to be present during a general shooter incident. In many cases, the general shooter would shoot at the nonpolice target, and then flee the scene. They would avoid arrest this way because no police officer would be immediately present at the scene to pursue the shooter. Once the police did arrive, or the victim was found, the shooter, in many cases, was not present. This made it less likely for the police to have the ability to make an arrest during the incident because an investigative approach would be required. The PSS is at a greater probability of being arrested than the GSS because the PSS engages directly with police. The GSS had a very significant association with the drug trade being involved in the incident, and they were far more likely than the PSS incidents to have the drug trade involved in THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 33 their shooting incident. This is because there is a strong connection between the drug trade and violence with firearms. Addtionally, the drug trade is mainly composed of nonpolice individuals. This results in more general shooters in the drug trade. Firearms are used violently in the drug trade because many Canadian gang-related firearm homicides occur for drug trade reasons. Some of these reasons for gang-related firearm killings are for drug debts, the targeting a rival gang member, or a violent “turf war” between multiple gangs (Beattie, et al., 2018). Gang violence with firearms in the drug trade is also observable in the Canadian province of British Columbia (BC). In BC, gangs use firearms to gain control of the drug trade. These gangs want to dominate the drug trade because it is a very profitable illicit market (Heemskerk, & Davies, 2008). Any competition for drug profits would put a gang’s own operations at risk of reduced profit, so they will use firearms to protect their ability to make financial profits from selling drugs. Furthermore, being involved in drug-related gang activity increases a person’s chances of committing violence with firearms. When a BC gangster is involved in the drug trade, they are more likely to have firearms, and they are more likely to commit violence with firearms (Heemskerk, & Davies, 2008). Many cases in the GSS involved the drug trade, and all of them proved that they were willing to use firearms violently against others. There is further Canadian evidence that shows involvement in the drug trade is linked to shooters targeting nonpolice targets. Being involved in criminal activity, such as drug trafficking, greatly increases the risk of death by homicide, and firearm homicides in this field are prevalent. 89% of all Canadian homicide victims are involved in the crime business, crime security, or crime transportation, and killing others by firearm accounts for 87% of all Canadian gang related homicides. Furthermore, gang killings with firearms is also much higher than other areas of homicide because nongangrelated homicides use firearms 27% of the time (Beattie, et al., 2018). Criminals are willing to THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 34 use a firearm violently to better position themselves in the drug trade. The majority of killing methods gangs use is by firearm, and they do this at a greater rate than nongang firearm killings. The drug trade is a violent cycle as gang members commit violence with firearms, and they are also at risk to be victims themselves of shootings. The link between the drug trade and general shooters in not a new societal issue. A study by Oscapella (1998), found that firearm related violence is used to regulate the drug trade by securing drug shipments, to collect drug debt, to resolve drug trade disputes, to intimidate competitors or customers, to kill informants, or to kill competitors. The drug trade is linked to general shooters because rival gangs use violence to supress one another, and gangs will use firearms to secure their power. There are reasons that the PSS is involved in the drug trade at far lower rates than the GSS. The drug trade was not associated with the PSS because a criminal operating as a general shooter would want to avoid a police officer while engaging in their more common activity of the drug trade. This is because a police officer is a major risk to the general shooters involved in the drug trade. For example, if a drug dealer is caught by a police officer, they may be arrested, have their drugs confiscated, and they will lose time to sell drugs in the future because they will be engaged in the criminal justice process. It is important for general shooters to avoid police for the utilitarian purposes of not losing their drugs, and to not loose the time to sell drugs in the future. The general shooters also want to avoid police because the consequences are arrest are not favourable as their freedoms would be reduced. This is why the GSS did not engage with the police as often, and reduced engagement resulted in lesser adversarial engagements with the police. The drug trade was not significant for the police shooter so it was less of a contribution to their violence with firearms when compared to the GSS. Rather, the police shooters had adult trauma as a significant factor that contributed to their shooting. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 35 There was a marginally significant association between the police shooter and experiencing a traumatic event in adulthood. The Social Learning Theory (SLT) and Biological Theory (BT) of trauma’s relationship to crime can explain why trauma was associated with the PSS. SLT states that aggression and violence are learned by individuals when they observe others model the behaviour. If the subject observes a traumatic experience that involves violence, the subject may learn to adopt violence against others. This is because the social learning experience can cause them to believe that violence is rewarding, able to be justified, and is acceptable (Neller & Fabian, 2006). If a police shooter is victimized by violent trauma in their adulthood, they may later believe that violence (such as by using a firearm) is the method to express their disturbed state. The BT is the next way to explain the PSS’s relation to trauma. Biological evidence has found trauma can increase violent behaviour of the subject in the future. Prior research found high levels of stress and potentially traumatic events reduces the serotonin production levels of the brain. Low serotonin levels have been associated with impulsive behaviour and aggression (Neller & Fabian, 2006). More recent research continues to support this concept found in previous studies. Research by Murrough et al. (2011) found there is a strong association between exposure to trauma and the reduction in serotonin receptors (type 1B receptor; meaning that serotonin levels are low because they are not received), and Grattan, et al. (2019) found trauma is associated with aggression towards others. The PSS has a strong level of adult trauma, and this has likely affected some of their serotonin in a negative way that favour aggression. Behaviours that trauma can led to include increased hostility, difficulty to regulate anger, and there are other emotional regulation problems, pessimistic thought processes, increased social detachment, more likely to incorrectly view others as malevolent. Many of these factors, such as anger, antisocial attitudes, cognitive distortions, and emotional dysregulation are THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 36 violent risk factors. It is important to note the majority of people who experience trauma do not commit future violence. However, research strongly suggests that trauma can directly predispose some people to commit future violence. Traumatic events may also contribute to a person’s development of alcohol misuse or emotional dysregulation, and those two factors are associated with violence (indirect contribution to violence) (Neller & Fabian, 2006). Trauma increased risk of committing violence and engaging in antisocial conduct (Neller & Fabian, 2006). Other chemicals such as cortisol and testosterone should be explored with their relation to trauma and aggression. Frequencies Discussion Both the PSS and GSS have a violent history (non-significant) in their top five frequencies. Having a violent history has been found to be associated with committing future violence. For instance, violence in childhood is associated with the later use of a weapon in a violent act (Michie, & Cooke, 2006) and a longitudinal study that focused on discharged psychiatric patients found a history of violence is associated with violent and criminal act (Beaudoin, 2019). Having a criminal record puts a person at a greater risk of committing a shooting. The PSS and GSS both have a criminal record in their top five frequencies (non-significant). The minority of Canadians have a criminal record because only 11.29% of Canadians have a criminal record (the discussion of the criminal record is in regards to any type of crime, such as a criminal record for drunk driving) (table 1O; Appendix O). However, those with criminal records are over represented in both shooting samples (over 60% of participants had a criminal record). Looking at violent offender particularly displays that they are also prone to reoffending. Canadian crime THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 37 statistics show that those convicted of a violent offence are prone to reoffend violently (Stewart, et al., 2019); • Those convicted of homicide (for men and women) will reoffend by committing a violent crime at a rate of 14.2% over five years • Those convicted of robbery (for men and women) will reoffend by committing a violent crime at a rate of 38.7% over five years • Those convicted of assault (for men and women) will reoffend by committing a violent crime at a rate of 36.9% over five years • Those convicted of other violent offences (for men and women) will reoffend by committing a violent crime at a rate of 27.2% over five years • Rates of violent reoffence after five years is slightly higher for men This reoffence rate may actually be higher because those incarcerated during the five-year follow up would have less opportunity to reoffended because they would be held in a secure facility. Having a criminal record display that the person is willing to commit criminal acts, and this research demonstrates that there is a likelihood of them continuing their behaviour. The drug use history is a factor of the shooter groups (non-significant) because people are willing to commit crimes to reengage with their drug use. This means if a person has used drugs, then they do not have access to the fund for the drugs, they may turn to other means, such as violent offences to gain the finances or physical drugs to continue the drug use. Among federal Canadian inmates, they reported that 39% of their gainful crimes (crimes to get something in return) were committed to gain either illicit drugs or a combination of drugs and alcohol THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 38 (Pernanen, et al., 2002). An example was found among the GSS, in which the shooter shot and killed the drug dealer so he could take the illicit drugs and use them. The majority of the GSS had a single parent (non-significant), and this contributed to their future risk of being a general shooter. Having a single parent disadvantages the child of the protective factors that two parents provide, and it poses some risk factors. Parents are important in the pro-socialization process of a child. A single parent will have less time to dedicate to their child’s pro-social development (such as teaching acceptable behaviour, how to respect others, and how to control unacceptable behaviour) when compared to having both parents (Singh & Kiran, 2012). This may result in the child developing characteristics that favour firearm related violence. For instance, male children from single parent homes are more aggressive than males from two parent homes (Rodríguez et al., 2013). Harmful attitudes that a child develops are also less likely to be detected early on in the single parent home because the child will be supervised less (Singh & Kiran, 2012). Delinquency that occurs during childhood due to the single parent home can increase the person’s risk of adulthood crime (National Institute of Justice, 2014), and there is a connection between childhood crime and later adulthood crime (Public Safety Canada, 2017). This occurs through the process known as the continuity of criminality. The continuity of criminality occurs when a crime-involved youth continues their crimes into adulthood. Children that are especially at risk of adult crime are those who commit crime earlier on during their childhood, and those who commit violence during their childhood (National Institute of Justice, 2014). Single parent homes place children at a great risk of being involved in crime because of lower supervision rates and because two parents are more effective at pro-socializing a child. These children are then more likely to commit crime in their adulthood, and may be why there is THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 39 a high percentage of general shooters who are from single parent backgrounds during their childhood. In many cases, the father relation issue (non-significant) results in the single parent factor because the mother is left alone to raise the child. However, a greater context in regards to fatherless children, and adults with father relation issues can be discussed. First, having a father relation issue for young people can increase crime. Wong (2018) found that youth crime increases with Fatherless homes. This can be further supported by the reports from Lawson (2012). Lawson (2012) reported the Caribbean-Canadian and African-Canadian experience of Fatherless homes in Toronto. Even though this was not a quantitative assessment, it offers some detailed qualitative insight. For instance, a single mother reported that the children discuss their absent Fathers a lot, and that the fathers would be replaced by other factors such as gangs. Faith leaders in the community also stated that Fatherless homes result in a lack of role models, and many of the children turned to gangs to substitute their families. These gangs then lead them to be involved in crimes. Lawson (2012) also did a review of prior research that found fatherlessness is linked to an increase in incarceration rates, increases in violence, and an earlier onset of death. An adult who has father relation issues can also be at a greater risk of committing crime. This is because the father contributes to the parental social control that typically helps reduce their adult child crime (Johnson et al., 2011). When the father is not involved in the life of their adult child, that person will not gain the protective factors of the father, so crime is more likely to occur. This last area of discussion will be how the criminal can be charged. The research of the written charge sections revealed that there are many ways a judge can find a shooter guilty of a criminal offence. These types of offences varied in the degree of evidence required to convict the person, so it was typical practice to see the crown lawyers THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 40 attempt to have the accused convicted of an offence that required a greater intent (such as murder), but if that failed, they would typically have the accused convicted of an offence with a lower degree of intent (such as second-degree murder or discharging a firearm to prevent arrest) Conclusion Overall, the PSS and GSS are similar, but they have some differences. Statistically significant differences between the two groups were found in the variables of engaging in the drug trade currently, experiencing trauma in adulthood, and the two groups were different in their initial arrest outcomes. This research will greatly improve the academic knowledge of shooters who target the police or others. This is because there are many unknowns when comparing shooters who target the police to shooters who target others. The literature review established that there is no clear consensus if the two populations are similar or have major differences. This research was able to explore those areas. This study will help inform law enforcement personnel by helping them better understand shooters who target the police, and shooters who target others. This knowledge can be turned into actions that improve justice intuitions by helping them further develop the risk analysis of police shooters. The study findings can also be used to help assist criminal analysts or frontline workers in their risk analysis profiles of police and criminal shooters. By helping improve the knowledge of justice professionals, it is expected that the research implications will have a positive impact on law enforcement efforts. Additionally, this study will help fill in the research gaps in regards to comparing the characteristics of the PSS and the GSS. The researcher also had personal goals during this project. The research was looking to personally develop his research skills, and this did occur. For example, he was able to learn how to use and read statistical software with the guidance of THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 41 the research supervisor. For further reference in this research, view table 1P (Appendix P) for a full list of definitions. Limitations The current study does have limitations. These limitations include the study’s sample size, a gender imbalance, the type of shooters typically included in the CanLII database, time constraints, and the coding method. First, the small sample size of this study only allowed for the detection of large differences between the groups. The non-significant results may change if a larger sample size was used. A larger sample would have been more effective because smaller differences could have been detected. Gender was a limitation of this study because male shooters were over represented in this study (n= 51), and there was only one case of a female shooter in this study (R v L.R.F., 2005). The shooter profiles in CanLII are a limitation because it is almost always about shooters who survive. This made it difficult to analyze police shooters and general shooters who died (during the incident) because they cannot be prosecuted in court. However, the current study was able to have some level of analysis of shooters who were killed (R v Duro, 2017; R v Hennessey, 2009). The general shooter and police shooter survival rates are most likely lower than what the current reports because CanLII does not typically report on shooters who died during the incident. The time constraints of this research did not allow some variables that were detected (towards the end of the research period) to be analyzed. They were not analyzed because all the cases reviewed before their detection would have to be reviewed again. This study was unable to analyze the prevalence of negative peers (associates of the shooter who were a negative influence), family crime (if the criminal had family who were involved in crime), organized crime involvement (of the shooter), if the shooter was living with someone other than their biological parent (such as living with a grandparent). These factors THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 42 were not detected early on, so they may not be as prevalent as other detected factors. Unfortunately, they could not be analyzed due to time constraints. Another limitation was the coding method used to review the presence of the dependent variables. The researcher believes there are ways that this can be improved, and it may assume that something was not existent because it was not reported in the legal document. Interviews with the shooters and other relevant sources (such as their parole officer or lawyer) would probably give more certain results to if a dependent variable is present. This does not mean the coding method is completely flawed, as it does make accurate inferences, some reasonable assumptions, and it has safeguards built into it to help prevent wrongful assumptions. Research methods can always be improved, and the future research in this field can learn from these limitations. Lastly, there were limitations due to time constraints. The dependent variables (below) were not able to be analyzed in this study, and they should be reviewed in future research. These variables were not studied because they were detected later in the research process, and the PDFs could not be reviewed due to time constraints. The variables include: • the prevalence of negative peers (associates of the shooter who were a negative influence) • family crime (if the criminal had family who were involved in crime) • organized crime involvement (of the shooter) • if the shooter was living with someone other than their biological parent (such as living with a grandparent) Recommendations Police departments should evaluate if a two-partner police patrol system (either fully or for part of the working shift) will improve their response to shooting incidents. There is evidence THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 43 that supports both a single and two-partner police patrol system, and departments should analyze if it will enhance officer safety and if it will improve their responses to firearm incidents (Anderson & Dossetor, 2012). If tactically appropriate, police should resort to using semi-automatic rifles to engage police-shooters. The present study found that Canadian police are more likely to be targeted by a long-gun than a handgun, and they are shot at by long-guns more often than nonpolice individuals. Canadian police typically carry handguns. If the officer is shot at, the officer should resort to using a rifle as soon as possible, rather than their service pistol. The officer should only do this if it is the right tactical decision, it is safe to do so, and if there is time to do so. This is recommended because rifles are a more effective firearms than handguns. Rifles are more accurate than handguns, and rifles are capable of firing more powerful cartridge types (Kraemer, 2009). The claim that rifles are a more effective option is further supported by a firearm industry article as rifles are typically more powerful, have greater ammunition capacity, and they are more accurate (Douglas, 2020). A rifle will better protect the police officer because it is more effective for self-defence than a service pistol. Furthermore, if the criminal is already using a rifle, the police officer should have their rifle ready as this firearm will better match the criminal’s firearm’s capability. Police should be ready to use a trauma-informed approach to aid in de-escalation. A trauma-informed approach will help when police are dealing with shooters who have trauma (such as adulthood trauma). This is limited in its application because the shooter must be willing to communicate and police must be able use this in a way that does not compromise their safety and the safety of others (for instance, a police officer currently being shot at should shoot back at the criminal and find cover, rather than attempt to use a trauma-informed approach). A trauma- THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 44 informed approach may be used during a standoff, during communications, or during a postarrest of the shooter. A trauma-informed approach means to be able to understand the past trauma and the impacts it had on the person. This approach is based on safety and getting the person connections for help. Police are able to play an important role in the trauma-informed approach, and police who are competent in trauma are better prepared to deal with vulnerable populations (Jones, 2020). Jones (2020) recommends the trauma informed training course created by the Canadian Police Knowledge Network (Canadian Police Knowledge Network, n.d.). This course is recommended by the researcher for dealing with trauma of the shooters, and each police department should see how they can use this to potentially develop better in house trauma informed approaches. This recommendation appears to be of use. If it is implemented as one of the tools for dealing with the shooter populations, it will need to be assessed to see if it is worth continuing to implement. Involving drug treatment courts (DTCs) as part of the rehabilitation process of shooters who have drug use issues may be of use. However, they may not be admitted due to their violent offence. Public Safety Canada (2012) has found DTC reduce reoffence rates from 4%-8%, though it is uncertain how this would apply to the shooter population specifically. The government should continue to fund initiatives that support family structures because general shooters have a high frequency of broken-family factors (e.g., single parent, father relation issue). There are government-supported programs that work with the community to enhance the family structure. Some people in the affected communities of broken families have claimed that improving the family structure has helped turn the youth away from gangs; which then helped address the violence with firearms (Lawson, 2012). However, this should be formally evaluated to ensure this is an effective practice. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 45 References Anderson, J., & Dossetor, K. (2012). First-response police officers working in single person patrols: A literature review. Australian Government. https://www.publicsafety.gc.ca/lbrr/archives/cnmcs-plcng/cn36352-eng.pdf Beattie, S., David, J., & Roy, J. (2018, November 21). Homicide in Canada, 2017. Statistics Canada. https://www150.statcan.gc.ca/n1/pub/85-002-x/2018001/article/54980-eng.htm Beaudoin, M., Potvin, S., Dellazizzo, L., Luigi, M., Giguère, C. E., & Dumais, A. (2019). Trajectories of dynamic risk factors as predictors of violence and criminality in patients discharged from mental health services: a longitudinal study using growth mixture modeling. Frontiers in psychiatry, 10, 301. Blackstone, A. (2012). Principles of sociological inquiry– Qualitative and quantitative methods. Saylor Foundation. https://open.umn.edu/opentextbooks/textbooks/139 Business Vancouver. (2016, January 12). Profile of Andrew Petter, President and ViceChancellor. Osgoode Digital Commons. https://digitalcommons.osgoode.yorku.ca/cgi/viewcontent.cgi?referer=&httpsredir=1&artic le=1166&context=media_mentions Canadian Legal Information Institute. (n.d.) Homepage. https://www.canlii.org/en/ Canadian Police Knowledge Network. (n.d.). Using a trauma-informed approach. https://www.cpkn.ca/en/course/using-a-trauma-informed-approach/ Cohen, I., & Garis, L. (2018, August). Determinants of injury and death in Canadian police officers. University of the Fraser Valley. https://cjr.ufv.ca/wp- THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 46 content/uploads/2018/09/Determinants-of-Injury-and-Death-in-Police-Officers-August2018.pdf Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Creswell, J.W., & Creswell, J.D. (2018). Research design (5th ed.). Sage Publications. Douglas, R. (2020, March 4). Pistol vs. rifle vs. shotgun: What's the difference? Bushnell. https://www.bushnell.com/expert-advice/bu-blog-pistol-vs-rifle-vs-shotgun-whats-thedifference.html Dunn, S. (2010, October 26). Police officers murdered in the line of duty, 1961 to 2009. Statistics Canada. https://www150.statcan.gc.ca/n1/pub/85-002-x/2010003/article/11354eng.htm Field, A. (2009). Discovering statistics using SPSS (3rd ed.). Thousand Oaks, CA: SAGE Publications Inc. Grattan, R.E., Lara, N., Botello, R.M., Tryon, V.L., Maguire, A.M., Carter, C.S., Niendam, T.A. (2019). A history of trauma is associated with aggression, depression, non-suicidal selfinjury behavior, and suicide ideation in first-episode psychosis. Journal of Clinical Medicine, 8(7), 1082. 10.3390/jcm8071082 Gekoski, A., Gray, J. M., & Adler, J. R. (2012). What makes a homicide newsworthy? UK National Tabloid Newspaper Journalists tell all. British Journal of Criminology, 52(6), 1212–1232. https://doi.org/10.1093/bjc/azs047 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 47 Government of Canada. (n.d.). R.S.C., 1985, c. C-46. https://laws-lois.justice.gc.ca/eng/acts/c46/PITIndex.html Heemskerk, T., & Davies, E. (2008, November). A report on the illegal movement of firearms in British Columbia. https://www2.gov.bc.ca/assets/gov/law-crime-and-justice/criminaljustice/police/publications/independent/special-report-illegal-movement-firearms.pdf Issitt, M. L. (2021). Shooting of Police Officers, Dallas (2016). Salem Press Encyclopedia. Justice Institute of British Columbia. (n.d.). Research Policies & Procedures. https://www.jibc.ca/applied-research-graduate-studies/research-policies-procedures Johnson, W. L., Giordano, P. C., Manning, W. D., & Longmore, M. A. (2011). Parent-child relations and offending during young adulthood. Journal of Youth and Adolescence, 40(7), 786–799. https://doi.org/10.1007/s10964-010-9591-9 Jones, D.J. (2020). Pandemic policing: Highlighting the need for trauma-informed services during and beyond the COVID-19 crisis. Journal of Community Safety and Well-Being, 5(2), 69-72. https://doi.org/10.35502/jcswb.129 Kraemer, J. (2009). A basic firearm tutorial. University of Utah. https://webpath.med.utah.edu/TUTORIAL/GUNS/GUNANAT.html Lawson, E. (2012). Single mothers, absentee fathers, and gun violence in Toronto: A contextual interpretation. Women’s Studies, 41(7), 805–828. https://doi.org/10.1080/00497878.2012.707903 Logan, M. H. (2021, November 29). Firearm access by at-risk people. Blue Line. https://www.blueline.ca/firearm-access-by-at-risk-people/ THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 48 McAleese, S., & Latimer, C. (2017). Reforming the criminal records act [Ph.D. Student Paper]. John Howard Society of Canada. https://johnhoward.ca/wpcontent/uploads/2017/04/Criminal-Records-Act-Reform-Final-1.pdf Michie, C., & Cooke, D.J. (2006). The structure of violent behavior: A hierarchical model. Criminal Justice and Behavior, 33(6), 706-737. https://doi.org/10.1177%2F0093854806288941 Moreau, G. (2021, July 27). Police-reported crime statistics in Canada, 2020. Statistics Canada https://www150.statcan.gc.ca/n1/en/pub/85-002-x/2021001/article/00013eng.pdf?st=TNJwyeo7 Mulligan, L., Axford, M., & Solecki, A. (2016, November 23). Homicide in Canada, 2015. Statistics Canada. https://www150.statcan.gc.ca/n1/en/pub/85-002x/2016001/article/14668-eng.pdf?st=emEJsKeC Murrough, J.W., Czermak, C., Henry, S., Nabulsi, N., Gallezot, J., Gueorguieva, R., PlanetaWilson, B., Krystal, J.H., Neumaier, J.F., Huang, Y., Ding, Y., Carson, R.E., & Neumeister, A. (2011). The effect of early trauma exposure on serotonin type 1b receptor expression revealed by reduced selective radioligand binding. Arch Gen Psychiatry, 68(9), 892–900. 10.1001/archgenpsychiatry.2011.91 National Institute of Justice. (2014, March 10). From youth justice involvement to young adult offending. https://nij.ojp.gov/topics/articles/youth-justice-involvement-young-adultoffending Neller, D.J. & Fabian, J.M. (2006). Trauma and its contribution to violent behaviour. Centre for Crime and Justice Studies, 66(1), 6-7. 10.1080/09627250608553387 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 49 Oscapella, E. (1998, July). The relationship between illegal drugs and firearms. Department of Justice Canada. http://publications.gc.ca/collections/collection_2011/jus/J3-8-1998-10eng.pdf Pannucci, C. J., & Wilkins, E. G. (2010). Identifying and avoiding bias in research. Plastic and Reconstructive Surgery, 126(2), 619–625. https://doi.org/10.1097/prs.0b013e3181de24bc Pernanen, K., Cousineau, M., Brochu, S., & Sun, F. (2002, April). Proportions of crimes associated with alcohol and other drugs in Canada. Canadian Centre on Substance Abuse. https://www.ccsa.ca/sites/default/files/2019-04/ccsa-009105-2002.pdf Public Safety Canada. (2017). Transitions from juvenile delinquency to young adult offending: a review of canadian and international evidence. https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/2017-s014/index-en.aspx Public Safety Canada. (2016, February). Corrections and conditional release statistical overview 2015. https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/ccrso-2015/ccrso-2015-en.pdf Public Safety Canada. (2012, November). "What Works" in Drug Treatment Courts. https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/wtwrks-crts/index-en.aspx Rodríguez, M.A., del Barrio, V., Carrasco, M. (2013). Physical and verbal aggression in single parent families versus two parent families: The moderator effect of the children's gender. Revista de psicopatología y psicología clínica 18(2),119-12. https://www.researchgate.net/publication/286519235_Physical_and_verbal_aggression_in_ single_parent_families_versus_two_parent_families_The_moderator_effect_of_the_childr en's_gender THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 50 R v Baugh (2022), 2021 CanLII 8408 (ONSC). https://canlii.ca/t/jm04l R v Crockwell (2022), 2013 CanLII 8675 (NL SC). https://canlii.ca/t/fw903 R v Duro (2022), 2017 CanLII 4573 (ONSC). https://canlii.ca/t/h54qn R. v. Fobister (2022), 2009 CanLII 31987 (ON SC). https://canlii.ca/t/242q8 R. v Hennessey (2022), 2009 CanLII 60 (ABQB). https://canlii.ca/t/fqj0h R v Kakfwi (2022), 2018 CanLII 30 (NWTSC). https://canlii.ca/t/hv9jw R v L.R.F. (2022), 2005 CanLII 192 (NSSC). https://canlii.ca/t/1ljct R v Morrill (2022), 2016 CanLII 638 (ABQB). https://canlii.ca/t/gvn16 R v P.B. (2022), 2021 CanLII 8443 (ONSC) 8443. https://canlii.ca/t/jlh5p Schmalleger, F., & Volk, R. (2018). Canadian criminology today: Theories and applications (6th ed.). Pearson Canada. Singh, A., & Kiran, U.V. (2012). Effect of single parent family on child delinquency. International Journal of Science and Research, 3(9), 866-868. https://www.ijsr.net/archive/v3i9/U0VQMTQ0MA%3D%3D.pdf Statistics Canada. (2021a). Table 35-10-0069-01 Number of homicide victims, by method used to commit the homicide [Data table, 2011-2020]. https://doi.org/10.25318/3510006901-eng Statistics Canada. (2021b). Table 17-10-0009-01 Population estimates, quarterly [Data table]. https://doi.org/10.25318/1710000901-eng Statistics Canada. (2017). 2016 Census topic: Population and dwelling counts. https://www12.statcan.gc.ca/census-recensement/2016/rt-td/population-eng.cfm THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 51 Statistics Canada. (2014). Homicide in Canada, 2013. https://www150.statcan.gc.ca/n1/en/pub/85-002-x/2014001/article/14108eng.pdf?st=aEH1KcjI Statistics Canada. (2012). Homicide in Canada, 2011. https://www150.statcan.gc.ca/n1/en/pub/85-002-x/2012001/article/11738eng.pdf?st=SEbXBX7z Statistics Canada. (n.d.). Population. https://www150.statcan.gc.ca/n1/pub/12-581x/2015000/pop-eng.htm Stewart, L.A., Wilton, G., Baglole, S., & Miller, R. (2019, August). A comprehensive study of recidivism rates among Canadian federal offenders. Correctional Service of Canada. https://www.csc-scc.gc.ca/research/005008-r426-en.shtml#17 Wong, K. (2018, September). Single-parent families, economic disadvantage, and youth crime. https://www.brandonu.ca/research-connection/files/Wong-wl-pdf.pdf Zhang, T., & Quin, Y. (2012). The Economic Impact of Firearm-related Crime in Canada, 2008. Department of Justice Canada. https://www.justice.gc.ca/eng/rp-pr/csjsjc/crime/rr13_7/rr13_7.pdf THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 52 Appendix A The Distribution of Canadian Population and of the Study’s Participants These tables reflect the population distribution of Canada and of the study. This study aimed to match the population distribution of Canada. Table 1A Canadian Population Provincial and Territorial Distribution Desired Reflection Place (Province Population Percent of Desired Cases in Case Percent of Canadian the Study (26 the Study Population per group) 519,664 1.3% 1 3.84% 161,514 0.42% 1 3.84% Nova Scotia 982,012 2.57% 1 3.84% New Brunswick 784,156 2.05% 1 3.84% Quebec 8,579,370 22.53% 4 15.38% Ontario 14,759,431 38.77% 7 26.92% Manitoba 1,381,459 3.62% 1 3.84% Saskatchewan 1,178,971 3.09% 1 3.84% Alberta 4,431,454 11.64% 3 11.53% British 5,163,919 13.56% 3 11.53% or Territory) Newfoundland and Labrador Prince Edward Island Columbia THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Yukon 42,344 0.11% 1 3.84% Northwest 45,323 0.11% 1 3.84% 39,255 0.10% 1 3.84% 53 Territories Nunavut Note. This table demonstrates population estimates of Canada for the first quarter of 2021 (Statistics Canada, 2021b). Table 2A Canadian Population Provincial and Territorial Distribution Actual Reflection Place Population Percent of Representation Police (Province or Canadian of Both Territory) Population Samples Police Shooter Shooter Shooter Shooter Sample Sample Sample Sample (26 Percent (26 Percent total) Newfoundland 519,664 General General total) 1.3% 3.85% 1 3.85% 1 3.85% 0.42% 1.93% 0 0% 1 3.85% and Labrador Prince Edward 161,514 Island Nova Scotia 982,012 2.57% 3.85% 1 3.85% 1 3.85% New 784,156 2.05% 3.85% 1 3.85% 1 3.85% Quebec 8,579,370 22.53% 9.62% 4 15.38% 1 3.85% Ontario 14,759,431 38.77% 28.85% 5 19.23% 10 38.46% Brunswick THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 54 Manitoba 1,381,459 3.62% 3.85% 1 3.85% 1 3.85% Saskatchewan 1,178,971 3.09% 7.7% 3 11.54% 1 3.85% Alberta 4,431,454 11.64% 9.62% 2 7.69% 3 11.54% British 5,163,919 13.56% 15.39% 5 19.23% 3 11.54% Yukon 42,344 0.11% 3.85% 1 3.85% 1 3.85% Northwest 45,323 0.11% 3.85% 1 3.85% 1 3.85% 39,255 0.10% 3.85% 1 3.85% 1 3.85% Columbia Territories Nunavut Note. This table demonstrates how close the current studies population reflects geographic spread the population of Canada THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 55 Appendix B Search Types and Revisions These tables reflect the population distribution of Canada and of the study. This study aimed to match the population distribution of Canada. Table 1B Initial and Revised CanLII Search Database Search Term Number of Hits CanLII [GSS initial search] "firearm" 14,499 and "convicted" and "discharging" or "intent to cause bodily harm" or "violent" or "violence" CanLII [GSS second initial search] 77 "firearm" and "discharging" and "intent to cause bodily harm" CanLII [GSS revised search] "firearm" or "gun" or "handgun" or "pistol" or "rifle" or "shotgun" and "discharg*" or "shot" or 577 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS "fired at" and "convict*" or "guilty" and "intent to cause bodily harm" or "reckless endangerment" or "intent to wound" CanLII [PSS initial search] 2,379 discharged a firearm at police "convict*" "shot" CanLII [PSS second initial search] 77 discharged a firearm "at police" "convict*" "shot" CanLII [PSS revised search] "shot at police" or "shooting at police" or "shot police" or "shot detective" or "shot at corporal" or "shot corporal" or "shot at constable" or "shot constable" or "shot at officer" or "shot officer" or "fired at officer" or "firing at police" or "firing at corporal" or "endangering police" or "discharg* at police" and 116 56 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS firearm or "gun" or "handgun" or "pistol" or "rifle" or "shotgun" and "convict*" or "guilty" Note. This table demonstrates the initial and revised search terms of the PSS and GSS. 57 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 58 Appendix C Coding Criteria Method This is the coding criteria method that was used to determine if a variable was present, not present, or missing. If the factors are mentioned anywhere in the document they will be counted, otherwise if the applicable section/relevant area is discussed without the factor being mentioned, it will be assumed that factor is not present (no). However, this assumption does not apply to everything. There are some factors that do require more explicit mentions, otherwise they will be 99. Reference the operationalization sheet for further inclusion detail on each factor if required. Section: Facts of the event (when the case discuses the fact of the event in detail) *Assumed factor is “no” if not mentioned and this section (facts of the event) exists. If the facts of the event are not mentioned, then it is 99 (missing value) Partner violence current Mental health current Partner trouble current Family trouble current Dependent trouble current Family violence current Alcohol use current Extended family trouble current Drug use current Extended family violence current Addiction current Nonfamily trouble current Suicide risk current THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 59 Section: Circumstances of the offender (or Gladue Factor section, “the accused” section, or if there is a section that discusses the accused’s background) *If the circumstance of the offender section is mentioned and the factor is not present, it is assumed “no.” If the circumstances of the offender are not mentioned, then the factor is 99 (missing value) Alcohol use history Violent history Drug use history Prior threat to law enforcement (or no criminal record) Addiction history Traumatic event (not childhood) Suicide risk history Associate violence Mental health history Hunter Antisocial emotions Traumatic event (not childhood) Specific Areas relevant to a Certain Factor *The brackets section is what must be mentioned for it to be assumed that the factor is not present (“no”). If no relevant section or discussion to the factor is made, it is a missing value. Childhood trauma (if childhood (including any part of child up brining) mentioned) Childhood inadequate caregiver (if childhood, parent) Childhood school trouble (if childhood (including any part of child up brining) or school mentioned) THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 60 Childhood crime (if criminal record or childhood (including any part of child up brining) mentioned) Childhood divorce (must be explicitly mentioned, or assumed no divorce if parents are supportive during childhood and later on) Single parent (must be explicitly mentioned, or raised by one parent, implying they are single) Firearms licence (must be explicitly mentioned if they have one or not, or violate the relevant criminal code, or if they are violating a firearms prohibition during the offence) Parental drug use (if parent or person raising them mentioned) Family trouble history (if any part of family is mentioned and the circumstances of offender disused or the family is discussed in this regard) Family violence history (if any part of family mentioned and the circumstances of offender disused or or the family is discussed in this regard) extended family violence history (if any part of extended family mentioned and the circumstances of offender disused or the extended family is discussed in this regard) extended family trouble history (if any part of extended family mentioned and the circumstances of offender disused or the extended family is discussed in this regard) nonfamily trouble history (if any nonfamily relationships are discussed and no issues are found) Intergenerational trauma (if Indigenous and impacted by the operationalized definition criteria) Partner violence history (if partner(s) exists) Partner trouble history (if partner(s) exists) THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 61 Dependent trouble history (if has kids) Unemployment history (if any work is discussed) Unemployment current (if any work is discussed) Marital status (this usually requires an explicit statement, but if the criminal’s social relationships are discussed, and no partner is presented, it will be assumed “single.” Assumed common law if boyfriend/girl friend) Education (must be mentioned) Father Relationship Issue (either their circumstances are discussed and the father is mentioned with no issues, or it is explicitly stated) Mother Relationship Issue (either their circumstances are discussed and the mother is mentioned with no issues, or it is explicitly stated) Foster care (assumed no if circumstances discussed or that both the person’s parents are involved in raising them or in their life) Assuming no if not mentioned anywhere in the document (does not require a specific section, but the court case document must have detail on the facts of the event) (See R. v. Duro, 2017 ONSC 4573 (CanLII) for an example of a case that lacks discussing any detail other than minimal inclusion) Court case document that are too vague (extremely minimal information) will have these factors as missing values (99s) Drug trade current Sleep Deprivation THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Drug trade history Racism Aggressive music obsession Misogyny Violent video game addiction Poverty Firearm obsession Community partner violence Community violence with firearms Corrupt government current Community drug use Associate violence history 62 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 63 Appendix D Coding Criteria Method This shows how the variables were grouped into categories by the researcher. This was done to improve the review process of the variables. The grouping of the variables was based on their relevance to one another. Table 1D Categories of Variables Category Variable generated Case documentation (6 variables) Case identity Case name Case link Case type Sample type Code conviction Circumstance of incident (19 variables) Jurisdiction Offence year RC Reason police called RC Reason nonpolice targeted Incident mins THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Who shot first Police shoot Shots fired criminal Shots fired police Shots hit criminal Shots hit police Shots hit nonpolice Police killed Criminal killed Nonpolice killed RC Resolvent Firearm type Drug trade Associate violence history Demographic (7 variables) Gender Age Ethnicity Ethnicity mixed Ethnicity other Education Marital status 64 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Childhood (7 variables) Foster care Single parent Childhood inadequate caregiver Childhood trauma Childhood divorce Childhood school trouble Childhood crime Relationship (19 variables) Partner violence current Partner violence history Partner trouble current Partner trouble history Dependent trouble current Dependent trouble history Father relation issue Mother relation issue Parental drug use Family trouble current Family trouble history Family violence current Family violence history Extended family trouble current 65 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Extended family trouble history Extended family violence current extended family violence history Nonfamily trouble current Nonfamily trouble history Substance use (6 variables) Alcohol use current Alcohol use history Drug use current Drug use history Addiction current Addiction history Mental health (6 variables) Suicide risk current Suicide risk history Mental health history Mental health current Sleep deprivation Antisocial emotions Characteristics of the shooter (13 variables) Corrupt government view Traumatic event Violent history 66 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Firearm licence Firearm obsession Hunter Criminal record Aggressive music obsession Violent video game addiction Prior threat to law enforcement Misogyny Racism Intergenerational trauma Socioeconomic (3 variables) Unemployment current Unemployment history Poverty Community factors (3 variables) Community offences with firearms Community drug use Community partner violence Note. This table shows the categories, the number of variables it applies to, and the type of variables in the category. 67 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 68 Appendix E Top Five Frequencies of the PSS and GSS These tables list the five most common frequencies of the samples. There may be more than five frequencies on the table if they have the same percentage. Table 1E Top Five Frequencies of the PSS Highest Percent Variable Name Percent of “yes” Rank Total applicable responses (n= yes/no responses) 1 Violent history 66.7% 18 2 Drug use history 61.1% 18 2 Criminal record 61.1% 18 3 Intergenerational 57.1% 7 trauma 3 Unemployment 57.1% 14 56.3% 16 55.6% 18 current 4 Unemployment history 5 Alcohol use history THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 69 Note. This chart demonstrates the highest percent scores of when a variable is present in the PSS Table 2E Top Five Frequencies of the GSS Highest Percent Variable Name Percent of “yes” Rank Total applicable responses (n= yes/no responses) 1 Childhood 68.8% 16 trauma 1 Single parent 68.8% 16 2 Father relation 66.7% 15 issue 3 Criminal record 64.7% 17 4 Violent history 61.1% 18 5 Unemployment 58.3% 12 history Note. This chart demonstrates the highest percent scores of when a variable is present in the GSS Table 3E Top Five Frequencies of the PSS and GSS Combined Highest Percent Rank Variable Name Percent of “yes” Total applicable responses (n= yes/no THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS responses) 1 Violent history 63.9% 36 2 Criminal record 62.9% 35 3 Single parent 59.4% 30 4 Childhood trauma 57.75% 31 5 Unemployment 53.55% 24 current Note. This chart demonstrates the highest percent scores of when a variable is present in the combined samples of the GSS and PSS 70 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Appendix F Raw Reporting Numbers of the PSS and GSS These tables discuss the raw reporting numbers of the PSS and GSS. Table 1F Top Five Raw Reported Factors of the PSS Most number Variable Name reported “yes” Number reported Total applicable “yes” rank responses (n= yes/no responses) 1 Violent history 12 18 2 Drug use history 11 18 2 Criminal record 11 18 3 Alcohol use 10 18 10 19 9 16 history 3 Mental health history 4 Unemployment history 4 Traumatic event 9 18 5 Unemployment 8 14 current 71 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 5 Childhood crime 8 16 5 Alcohol use 8 24 8 24 72 current 5 Mental health current Note. This chart demonstrates the scores that have the highest number of reported “yes in the PSS. Table 2F Top Five Raw Reported Factors of the GSS Most number Variable Name reported “yes” Number reported Total applicable “yes” rank responses (n= yes/no responses) 1 Drug trade 13 23 11 16 current 2 Childhood trauma 2 Single parent 11 16 2 Criminal record 11 17 2 Violent history 11 18 3 Father relation 10 15 issue THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 3 Drug trade 10 23 7 12 7 15 73 history 4 Unemployment history 4 Childhood school trouble 4 Drug use history 7 19 5 Childhood crime 6 15 5 Family trouble 6 16 6 25 6 25 history 5 Alcohol use current 5 Community offences with firearms Note. This chart demonstrates the scores that have the highest number of reported “yes in the GSS. Table 3F Lowest Five Raw Reported Factors of the PSS Lowest number reported “yes” rank Variable Name Number reported Total applicable “yes” responses (n= yes/no THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS responses) 1 Dependent 0 21 0 10 0 14 0 25 0 4 0 12 0 20 trouble current 1 Dependent trouble history 1 Family violence history 1 Extended family trouble current 1 Extended family violence history 1 Nonfamily trouble history 1 Associate violence history 1 Firearm licence 0 14 1 Community 0 24 partner violence 1 Misogyny 0 24 1 Racism 0 24 2 Poverty 1 24 2 Firearm 1 24 obsession 74 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 2 Aggressive 1 24 1 24 1 19 1 5 2 25 2 24 music obsession 2 Violent video game addiction 2 Prior threat to law enforcement 2 Extended family trouble history 3 Extended family violence current 3 Family trouble current 3 Foster care 2 20 4 Nonfamily 3 25 3 24 3 24 3 24 3 21 trouble current 4 Sleep deprivation 4 Corrupt government view 4 Family violence current 4 Partner violence 75 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS current 4 Partner trouble 3 21 3 13 4 25 4 24 Community drug 4 24 current 4 Childhood divorce 5 Drug trade current 5 Suicide risk current 5 use 5 Suicide risk 4 18 4 18 history 5 Antisocial emotions 5 Hunter 4 16 5 Partner violence 4 15 4 15 4 14 history 5 Childhood inadequate caregiver 5 Family trouble history 76 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 5 Intergenerational 4 7 trauma This chart demonstrates the ranking of the lowest number of “yes” for the PSS. Table 4F Lowest Five Raw Reported Factors of the GSS Lowest number Variable Name reported “yes” Number reported Total applicable “yes” rank responses (n= yes/no responses) 1 Aggressive 0 25 0 25 0 12 0 26 0 26 0 10 0 15 music obsession 1 Dependent trouble current 1 Dependent trouble history 1 Extended family trouble current 1 Extended family violence current 1 Extended family violence history 1 Firearm licence 77 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 1 Firearm 0 24 0 26 0 25 1 26 obsession 1 Sleep deprivation 1 Violent videogame addiction 2 Corrupt government view 2 Misogyny 1 26 2 Suicide risk 1 26 Community drug 1 25 current 2 use 2 Community 1 25 1 17 1 16 1 12 partner violence 2 Associate violence history 2 Family violence history 2 Partner violence history 78 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 2 Extended family 1 10 Intergenerational 1 3 trouble history 2 trauma 3 Family violence 2 26 2 26 current 3 Partner violence current 3 Racism 2 26 3 Drug use current 2 25 3 Prior threat to 2 21 law enforcement 3 Foster care 2 20 3 Childhood 2 17 2 17 2 16 3 26 3 26 inadequate caregiver 3 Nonfamily trouble history 3 Parental drug use 4 Family trouble current 4 Partner trouble 79 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 80 current 4 Addiction 3 25 current 4 Poverty 3 25 4 Hunter 3 18 4 Suicide risk 3 18 history 4 Traumatic event 3 18 4 Partner trouble 3 12 4 24 4 18 4 11 history 5 Mental health current 5 Addiction history 5 Childhood divorce Note. This chart demonstrates the ranking of the lowest number of “yes” for the GSS. Table 5F Highest Five Raw Reported Factors of both the PSS and GSS combined Highest number reported “yes” rank Variable Name Number reported Total applicable “yes” responses (n= yes/no responses) THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 1 Violent history 23 36 2 Criminal record 22 35 3 Single parent 18 30 3 Childhood trauma 18 31 3 Drug use history 18 37 4 Drug trade current 17 48 4 Drug trade history 17 45 5 Unemployment 16 28 81 history This chart demonstrates the ranking of the highest number of “yes” in the combined samples of the GSS and PSS Table 6F Lowest Five Raw Reported Factors of the PSS and GSS Combined Lowest number Variable Name reported “yes” rank Number reported Total applicable “yes” responses (n= yes/no responses) 1 Dependent trouble 0 46 0 22 0 51 current 1 Dependent trouble history 1 Extended family trouble current THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 1 Extended family 0 14 violence history 1 Firearm licence 0 29 2 Aggressive music 1 49 1 37 1 49 1 30 obsession 2 Associate violent history 2 Community partner violence 2 Family violence history 2 Firearm obsession 1 48 2 Misogyny 1 50 2 Violent video game 1 49 2 15 2 51 2 29 addiction 3 Extended family trouble history 3 Extended family violence current 3 Nonfamily trouble history 3 Racism 2 50 4 Sleep deprivation 3 50 82 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 4 Prior threat to law 3 40 enforcement 5 Foster care 4 40 5 Poverty 4 49 Note. This chart demonstrates the ranking of the lowest number of “yes” in the combined samples of the GSS and PSS 83 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 84 Appendix G Demographic Statistical Tests This demonstrates the statistical tests that were conducted to see if there were any differences between the two samples in their demographic variables. There were no significant demographic differences. Table 1G Testing for Demographic Differences Demographic Gender Ethnicity Education Age Martial Status Statistical Test Type Fisher’s Exact Test FisherFreemanHalton Exact Test FisherFreemanHalton Exact Test MannWhitney U Test FisherFreemanHalton Exact Test Exact Sig. (2Statistically sided) (p value) Significant 1 No 0.414 No 0.871 No 0.165 No 0.78 No Note. This chart demonstrates the statistical tests used to detect any differences among the two sample groups compared. There were no significant demographic differences between the two sample groups. If any statistical score was p<0.05 then it would be deemed significant. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Appendix H Types of Shooting Convictions This demonstrates type of conviction a criminal shooter can be convicted of. Table 1H Types of Convictions for Criminal Shooting Conviction Type • Murder • Second degree murder • Attempted murder • Discharging a firearm to prevent arrest • The unauthorized possession of a firearm • Not having a firearm licence while in possession of a firearm • Possession of a firearm knowing it was unauthorized due to a court order • Possession of a weapon for a purpose dangerous to the public peace • Discharging a firearm with intent to wound • The reckless discharge of a firearm • Assault with a firearm • Using a firearm during an offence • Unlawfully pointing a firearm • The careless use of a firearm • Having a firearm that was obtained by the commission of an offence 85 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS • 86 The transferring of a firearm knowing it was not authorized to do so Note. This is a list of the types of convictions a criminal shooter can be convicted of among the PSS and the GSS. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 87 Appendix I Shooting Incident Factors This area discusses the outcomes that occur during the shooting incident. Table 1I Group Shooting Engagement, Accuracy, Death, and Survival Rate Police Variable Name Percentage n (%) Police Shooter General Shooter Nonpolice Target Percentage n Percentage n Percentage n (%) (%) (%) Shot first 4 25 96 25 100 26 0 26 Likelihood to shoot in 61.54 26 100 26 100 26 NA NA 46.15 5 26.67 19 19.5 20 NA NA 15.38 26 7.69 26 0 26 20 25 84.62 26 92.31 26 100 26 80 25 incident Shot accuracy (shots that hit target) Chance of death of at least one group member Survival rate of group member Note. This table looks four groups; the police, the police shooter, the general shooter, and the nonpolice target. It compares shooting related factors between the groups. It was not considered if nonpolice targets shot back. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 88 Table 2I Shots Fired by Groups Police Shooter General Shooter n Number n Number n Total shots fired by the group 13 5 45 19 74 22 Average number of shots 5 2.37 19 3.36 22 Variable Name Police Number 2.6 fired per incident Note. This chart looks at three groups; the police, the police shooter, and the general shooter. It considers the total shots fired by each group, and the average shots fired by each group per incident. Table 3I Long-guns and Pistols Used by Shooters Firearm Category Police Shooter General Shooter Percentage Percentage n n (%) (%) Rifle or Shotgun 61.90 13 45.45 10 Pistol 38.09 8 54.54 12 Note. This chart demonstrates the type of firearm used against the police by the PSS (n= 21), and against the nonpolice by the GSS (n=22). The homemade firearm and unknown firearms (that were not known if it was a long-gun or a pistol) were not included in this chart. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Appendix J Variable Overview This table demonstrates the percentages and raw numbers of each variable. It is in alphabetical order. Table 1J Variable Overview Variable Name Police Shooter Sample General Shooter Sample Percentage (%) Percentage N n (%) Addiction current 29.2 24 12 25 Addiction history 41.2 17 22.2 18 Aggressive music obsession 4.2 24 0 25 Alcohol use current 33.3 24 24 25 Alcohol use history 55.6 18 27.8 18 Antisocial emotions 22.2 22 26.3 19 Associate violent history 0 20 5.9 17 Childhood crime 50 16 40 15 Childhood divorce 23.1 13 36.4 11 Childhood inadequate 26.7 15 11.8 17 46.7 15 46.7 15 caregiver Childhood school trouble 89 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Childhood trauma 46.7 15 68.8 16 Community drug use 16.7 24 4 25 Community offences with 20.8 24 24 25 Community partner violence 0 24 4 25 Corrupt government view 12.5 24 3.8 26 Criminal record 61.1 18 64.7 17 Dependent trouble current 0 21 0 25 Dependent trouble history 0 10 0 12 Drug trade current 16 25 56.5 23 Drug trade history 31.8 22 43.5 23 Drug use current 20.8 24 8 25 Drug use history 61.1 18 36.8 19 Extended family trouble 0 25 0 26 20 5 10 10 8 25 0 26 0 4 0 10 Family trouble current 8.3 24 11.5 26 Family trouble history 28.6 14 37.5 16 firearms current Extended family trouble history Extended family violence current Extended family violence history 90 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Family violence current 12.5 24 7.7 26 Family violence history 0 14 6.3 16 Father relation issue 35.7 14 66.7 15 Firearm licence 0 14 0 15 Firearm obsession 4.2 24 0 24 Foster care 10 20 10 20 Hunter 25 16 16.7 18 Intergenerational trauma 57.1 7 33.3 3 Mental health current 33.3 24 16.7 24 Mental health history 52.6 19 26.3 19 Misogyny 0 24 3.8 26 Mother relation issue 33.3 15 33.3 15 Nonfamily trouble current 12 25 20 25 Nonfamily trouble history 0 12 11.8 17 Parental drug use 33.3 15 12.5 16 Partner trouble current 14.3 21 11.5 26 Partner trouble history 40 15 25 12 Partner violence current 14.3 21 7.7 26 Partner violence history 26.7 15 8.3 12 Poverty 4.2 24 12 25 Prior threat to law 5.3 19 9.5 21 enforcement 91 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Racism 0 24 7.7 26 Single parent 50 14 68.8 16 Sleep deprivation 12.5 24 0 26 Suicide risk current 16.7 24 3.8 26 Suicide risk history 22.2 18 16.7 18 Traumatic event 50 18 16.7 18 Unemployment current 57.1 14 50 10 Unemployment history 56.3 16 58.3 12 Violent history 66.7 18 61.1 18 Violent video game addiction 4.2 24 0 25 92 Note. This table demonstrates the frequencies of factors relevant to the shooter samples. The “n” is based on the total “yes” and “no” responses, and the percent is the frequency of the “yes.” THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 93 Appendix K Testing Group Difference on Scale Dependent Variables This table demonstrates the outcomes of the Mann Whitney tests of the scale dependent variables. Table 1K Mann Whitney Analyses Testing Group Differences on Scale Dependent Variables Variable Mdn- GSS Mdn- PSS (median) (median) Age 23 Incident mins U z p 28 205.5 9.92 0.17 2 224 10 2553.17 0.33 Shots fired criminal 2 2 193 3.09 0.66 Shots fired police NA 0 No test No test No test Shots hit criminal NA 0.5 No test No test No test Shots hit police NA 0 No test No test No test Shots hit nonpolice NA 1 No test No test No test Police killed NA 0 No test No test No test Criminal killed 0 0 No test No test No test Nonpolice killed NA 0 No test No test No test Note. No test= No data from GSS so the test cannot be conducted. * p < 0.08, ** p < 0.05, *** p < 0.01, **** p < 0.001 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 94 Appendix L Testing Group Difference on Scale Dependent Variables This table demonstrates the outcomes of Chi-square tests on nominal dependent variables. Table 1L Chi-square analyses Testing Group Differences on Nominal Dependent Variables Variable x2 df p odds ratio Ethnicity 5.45 5 0.41 NA Gender 1.02 1 1 NA Marital status 1.13 2 0.78 NA Education 3.25 5 0.87 NA RC reason police called 6.77 6 0.29 NA RC reason nonpolice targeted NA NA NA NA Who shot first [if criminal shot 1.06 1 0.49 NA Police fired NA NA NA NA If criminal killed by police 2.08 1 0.49 NA Resolvent- suspect arrested 44.98 7 0.001**** 6.75 Resolvent- suspect fled 44.98 7 0.001**** PSS has zero so first] odds ratio cannot be THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS calculated Partner violence current 0.53 1 0.64 NA Partner violence history 1.49 1 0.34 NA Partner trouble current 0.08 1 1 NA Partner trouble history 0.68 1 0.68 NA Dependent trouble current No group did NA NA NA NA NA NA this Dependent trouble history No group did this Alcohol use current 0.52 1 0.54 NA Alcohol use history 2.9 1 0.18 NA Drug use current 1.65 1 0.25 NA Drug use history 2.18 1 0.19 NA Addiction Current 2.22 1 0.17 NA Addiction History 1.46 1 0.29 NA Suicide Current 2.28 1 0.18 NA Suicide History 0.18 1 1 NA Mental Health History 2.75 1 0.18 NA Mental Health Current 1.78 1 0.32 NA Sleep Deprivation 3.46 1 0.1 NA Corrupt Government View 1.27 1 0.34 NA Anti Social Emotions 0.08 1 1 NA Traumatic Event [adulthood] 4.5 1 0.075* 2.94 95 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Childhood Trauma 1.55 1 0.29 NA Childhood Inadequate Caregiver 1.16 1 0.38 NA Childhood divorce 0.51 1 0.66 NA Childhood school trouble 0 1 1 NA Childhood crime 0.31 1 0.72 NA Single parent 1.9 1 0.46 NA Father relation issue 2.78 1 0.14 NA Mother relation issue 0 1 1 NA Parental drug use 1.92 1 0.22 NA Foster care 0 1 1 NA Family trouble current 0.14 1 1 NA Family trouble history 0.27 1 0.71 NA Family violence current 0.32 1 0.66 NA Family violence history 0.91 1 1 NA Extended family trouble current No group did NA NA NA this Extended family trouble history 0.29 1 1 NA Extended family violence current 2.17 1 0.24 NA Extended family violence history No group did NA NA NA this Nonfamily trouble current 0.6 1 0.7 NA Nonfamily trouble history 1.52 1 0.5 NA Violent history 0.12 1 1 NA 96 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS Associate violence history 1.21 1 0.46 NA Intergenerational trauma 0.48 1 1 NA Unemployment current 0.12 1 1 NA Unemployment history 0.01 1 1 NA Poverty 0.01 1 0.61 NA Firearm licence No group did NA NA NA this Firearm obsession 1.02 1 1 NA Hunter 0.36 1 0.68 NA Firearm type 6.08 8 0.73 NA Criminal record 0.05 1 1 NA Drug trade current 8.6 1 0.006*** 3.56 Drug trade history 0.65 1 0.54 NA Aggressive music obsession 1.06 1 0.49 NA Violent video game addiction 1.06 1 0.49 NA Prior threat to law enforcement 0.26 1 1 NA Community offences with 0.07 1 1 NA Community drug use 2.14 1 0.19 NA Community partner violence 0.98 1 1 NA Misogyny 0.94 1 1 NA Racism 1.92 1 0.49 NA firearms 97 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 98 Note. NA due to being exclusive to one group, or because another group has zero so a comparison could not be conducted, or because the results were not significant so the odds ratio was not required to report. * p < 0.08, ** p < 0.05, *** p < 0.01, **** p < 0.001 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 99 Appendix M Testing Group Difference on Scale Dependent Variables This table demonstrates the outcomes of the Mann Whitney tests of the scale dependent variables. Table 1M Contingency Table: Resolvent Resolvent Type Outcome GSS PSS Total Suspect arrested Yes 2 14 16 No 22 12 34 Total 24 26 50 Yes 14.5 0.5 15 No 10.5 26.5 37 Total 25 27 52 Suspect Fled Scene Note. This table demonstrates the difference between the GSS and PSS outcomes in regards to being arrested of fleeing the scene. *The Woolf-Haldane Correction was made to this variable (original Yes for PSS was zero, add 0.5 to each number involved for the odds ratio to be conducted) Table 2M Contingency Table: Drug Trade Current THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS If the shooting GSS PSS Total Yes 13 4 17 No 10 21 21 Total 23 25 48 incident is drug trade related Note. This table demonstrates the GSS and PSS incident connection to the drug trade. Table 3M Contingency Table: Traumatic Event (Adulthood) If the shooter GSS PSS Total Yes 3 9 12 No 15 9 24 Total 18 18 36 experienced a traumatic event in adulthood Note. This table demonstrates the GSS and PSS outcomes when their adulthood trauma was measured. 100 THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 101 Appendix O Canadians with Criminal Records This table demonstrates the percent of Canadians who have a criminal record (for any crime). Table 1O Canadians with a Criminal Record (2009) Canadians with Criminal Canadian Population (2009) Record (2009) 3,800,000 Percent of Canadians who have a Criminal Record 33,628,571 11.29% Note. The Royal Canadian Mounted Police Criminal Records Division in 2009 reported 3.8 million Canadians had a criminal record (Public Safety Canada, 2016). The population of Canada in 2009 was 33,628,571 (Statistics Canada, n.d.). THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 102 Appendix P Variable Operationalization This table demonstrates the definitions of each variable in this study. Table 1P Variable Definitions Variable name Definition case_identity The unique identity code given to the case by the researcher case_name The legal title of the case as found on the CanLII website case_link The internet hyper link to the case case_type The type of court hearing (such as a trial of guilt or innocence, or a decision for sentencing) sample_type If the case is categorized in the general shooter sample or the police shooter sample jurisdiction The jurisdiction place (geographic area) of the offence offence_year The year of the shooting offence gender Whether the criminal is male or female age The age (in years) of the criminal at the time of the shooting ethnicity The ethnicity of the criminal ethnicity_mixed A combination of ethnicities ethnicity_other Another form of ethnicity that is not coded in ethnicity THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS code_conviction 103 The criminal code of Canada conviction(s) that resulted due to the criminal shooting. All charges in relation to the firearm(s) offences and the number is a criminal code statute. YCJA mentioned when applied reason_police_called The initial reason why police were called or responding to the incident RC_reason_police_called The initial reason the police responded to the incident. reason_nonpolice_targeted The reason the criminal shot at the nonpolice officer RC_reason_nonpolice_targeted The reason why the nonpolice person was shot at. incident_mins This measures how long the incident was in total; including the time that police were on site, or when they were first received the call for service, to when the incident was resolved (such as the subject is arrested) who_shot_first Whether the criminal or police fired the first shot police_shoot If the police shot their firearm at the criminal shots_fired_criminal The number of shots (bullets fired) that criminal shot at or towards the police or civilian target of the GSS. Does not include shots that had no target. Does not include rubber rounds, only real bullets. shots_fired_police The number of shots (bullets fired) that police shot at or towards the criminal. Does not include shots that had no target. Does not include rubber rounds, only real bullets. shots_hit_criminal The number of shots (fired by the police) that physically THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 104 impacted the criminal shots_hit_police The number of shots (fired by the criminal) that physically impacted the police shots_hit_nonpolice Nonpolice targets hit, usually in the context of a criminal shooter shooting at an associate criminal or non criminal person police_killed Number of police killed by the shooter with a firearm criminal_killed If the criminal was killed by the police nonpolice_killed The number of nonpolice people killed by the criminal's shots resolvement How the shooting incident was resolved RC_Resolvement How the incident was resolved (typically when the person was attested or if they fled. If they were in flight while the police were still looking for them, and then they were arrested, it counts as an arrest. education The formal educational level of the criminal marital_status The relationship status of the criminal partner_violence_current This occurs between current relationship partners and is any form of intimidation or emotional, physical, verbal, or psychological abuse. Current means on the day of the event partner_violence_history This occurs between current relationship partners and is any form of intimidation or emotional, physical, sexual, verbal, or psychological abuse. History means any day THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 105 prior to the offence date partner_trouble_current Any negative conduct that occurs in the criminal’s relationship with their partner (excludes partner violence variable). Current means on the day of the event partner_trouble_history Any negative conduct that occurs in the criminal’s relationship with their partner (excludes intimate partner violence). History means any day prior to the offence date dependent_trouble_current When the criminal is mistreating their child, stepchild, or a dependent. Child maltreatment. Current means on the day of the event dependent_trouble_history When the criminal is mistreating their child, stepchild, or a dependent. Child maltreatment. History means any day prior to the offence date alcohol_use_current If the criminal had consumed alcohol on the day of the offence. alcohol_use_history If the criminal had history of alcohol use (that was deemed problematic, was relevant, or if there was alcohol addiction). History means any day prior to the offence date drug_use_current If the criminal had consumed drugs or was under the influence of drugs during the offence (other than alcohol; the alcohol variable does not apply to this section). Current means on the day of the event drug_use_history If the criminal had a history of drug consumption (other THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 106 than alcohol) that was illicit drugs, problematic drug use, or the drug use was relevant. History means any day prior to the offence date addiction_current If the criminal had an addiction to a drug during the day of the offence (explicitly states an addiction, or going through the withdrawals of not using the drug). Current means on the day of the event. Includes diagnosed substance use disorder addiction_history If the criminal had an addiction (or was deemed to be dependent on the drug) to a drug prior to the day of the offence. History means any day prior to the offence date. Includes diagnosed substance use disorder. suicide_risk_current If there was indication that the criminal was suicidal during the day of the offence. Suicide includes ideation, attempts, or verbal statements of wanting to commit suicide. Current means on the day of the event suicide_risk_history If the criminal had a history of suicide prior to the offence day. Suicide includes ideation, attempts, or verbal statements of wanting to commit suicide. History means any day prior to the offence date mental_health_history Any mental wellness concern or condition that existed prior to the day of the offence, this does not include suicide or addiction variables. The psychiatric diagnosis history check THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 107 (after the crime) can also reveal this factor. History means any day prior mental_health_current Any mental wellness concern or condition that existed during the day of the offence, this does not include suicide or addiction variables. This condition can be diagnosed after the event as long as it would have existed during the crime. Current means on sleep_deprivation Any period of sleep deprivation that occurred during the days before, or the day of the incident corrupt_government_view Any view that believes the government (local, provincial, or federal), the justice system, or that Canadian laws are corrupt antisocial_emotions Any display of short temper, lack of remorse, or being manipulative towards others in the persons emotions. These emotions must be observed as part of the person's character (not due to drug use). Includes conduct disorder symptoms diagnosis traumatic_event Any traumatic event/predisposing (does not include childhood trauma) that had a major impact on the criminal. An event that occurred, or is perceived to have occurred as a mental health episode (mental event is being subject to a figure/environmental hazard). Does not include partner violence. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS childhood_trauma 108 Any childhood (age 17 years old or younger) event or series of events that had a major or lasting negative impact on the criminal. This may include but is not limited to physical, sexual, emotional, psychological, distressful, or traumatic events. This does not include divorce, but it can include witnessing family violence, parental fighting, witnessing parental substance abuse as a child. childhood_inadequate_caregiver Criminal is not appropriately cared during childhood (17 years of age of younger) for by their parents, caregivers, or other persons responsible for their upbringing. This can last for a certain period of time or for their whole childhood. childhood_divorce If the criminal’s parents divorced during the criminal's childhood (age 17 or younger, "childhood," "very young," or "child") childhood_school_trouble When the criminal, in childhood, receives school disciplinary action, skips school, quits school, gets into school fights, or is removed from school for causing issues childhood_crime This does not apply to young offenders (aged 17 or younger). When the criminal engaged in criminal activity in their childhood or had interactions with the youth justice system during their childhood. does not include personal drug use single_parent If the criminal was previously raised in a single parent THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 109 household at any point from 0-17 years of age father_relation_issue If the father was not consistently in the persons life or not involved in the shooter's upbringing or if it only states "raised by Mother," or "lived with Mother" (implies Father not actively involved or involved), or if Father died during criminal’s child mother_relation_issue If the mother was not consistently in the persons life or not involved in the shooter's upbringing "raised by Father", "lived with Mother" (implies Mother not actively involved or involved), or if Mother died during criminals’ childhood parental_drug_use If a parent of the criminal uses illicit drugs or abuse alcohol foster_care If criminal was in foster care or was adopted family_trouble_current When the criminal experiences major or reoccurring issues with their siblings or parents. Current means on the day of the event family_trouble_history When the criminal experienced a major or reoccurring issues with their siblings or parents prior to the offence date. This is in the context of family dispute of conflict and not abuse. History means any day prior to the offence date. Does not include childhood trauma, childhood inadequate caregiver, or divorce. family_violence_current If the criminal physically harmed, threatened, or verbally threatened a family member on the day of the offence, or THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 110 during the incident that the offence was part of. This occurs before the shooting. includes nieces/nephews. Does not include D/P family_violence_history If the criminal physically harmed, threatened, or verbally threatened a family member on any day prior to the offence day (Mother, Father, Sibling). Does not include dependent or partner violence extended_family_trouble_current If the criminal, on the offence day, has conflict (other than physical or violence related) towards an uncle, aunt, grandparent, or cousin extended_family_trouble_history If the criminal, prior to the offence day, has conflict (other than physical or violence related) towards an uncle, aunt, grandparent, or cousin extended_family_violence_current If the criminal, on the offence day, is violent (verbally or physically) towards an uncle, aunt, grandparent, or cousin extended_family_violence_history If the criminal, prior to the offence day, was violent (verbally or physically) towards an uncle, aunt, grandparent, or cousin nonfamily_trouble_current If the criminal, on the day of the offence, has a social or relational dispute with a person who is not a partner, family member, or relative (does not include violence committed against another as this is in violent history) nonfamily_trouble_history If the criminal, any day prior to the offence, has a social or THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 111 relational dispute with a person who is not a partner, family member, or relative (does not include violence committed against another as this is in violent history) violent_history If the criminal, prior to the offence day, had previous violent convictions, violent actions, or verbal threats against others associate_violence_history If the criminal physically harmed, threatened, or verbally threatened the victim (who is not a family member, dependent, or partner). Any day prior to the offence intergenerational_trauma Occurs when the criminal is Indigenous, and has been subjected to intergenerational or colonial trauma (a traumatic or major negative experience from another) that originated from an oppressive practice that targeted Indigenous peoples unemployment_current If the criminal did not have a job at the time of the offence. Current means on the day of the event. A current volunteer positions counts as employment. unemployment_history If the criminal had a history of frequently not being employed. History means any day prior to the offence date poverty If poverty was current or occurred the offence date and was relevant to the offender's current circumstances. Not having enough money to meet basic needs including food, clothing and shelter. THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS firearm_licence 112 If the criminal had a Canadian government issued firearms licence or, if licencing is unknown, but with a s. 91 charge of the criminal code (unauthorized possession; no firearms licence or no proper registration of the gun), s 109 violating prohibition firearm_obsession If the court finds the person was "obsessed with firearms" or if the person buys firearms to the point that it negatively affects their relationship with a family parent or spouse hunter If the criminal was preparing to go hunting or had a history of hunting (hunting for wild animals; not fishing) firearm_type The type of firearm used by the criminal shooter criminal_record If the criminal had a prior criminal record drug_trade_current If the criminal works in the drug trade or the circumstances of their engaged in are involved the selling, transportation, or possession of drugs/drug equipment in relation to the drug trade, or if the shooting is a drug trade related incident (debts) drug_trade_history If the criminal has a history of drug trade involvement or engagement. Any day prior to the offence aggressive_music_obsession If the criminal developed a habit of listening to aggressive music that the court found relevant to the criminal court discussion violent_video_game_addiction If the psychological or psychiatric assessment deemed the THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 113 criminal had an addiction to violent video games prior_threat_to_law_enforcement This occurs before the day of the offence. Any verbal or physical threats towards law enforcement (police, sheriff, corrections, bylaw, or jail guard), or any physically assaultive behaviour towards a law enforcement member community_offences_with_firear If the judge observed that the criminal's community had a ms concerning level of violence, of firearm offences, or an impacting level of violence on the criminal with firearms prior to the date of the offence community_drug_use If the community of the criminal or of the offence area suffers from drugs use. Almost every community does suffer this, but it is state it is relevant or it is an issue when it is mentioned in the court document community_partner_violence If the community of the criminal has a concerning rate of intimate partner violence that the judge views have some level of relevance to the criminal's offence misogyny If the criminal has a belief or attitude that expresses their personal sexist hatred of women (excludes intimate partner violence; i.e., partner violence variable). Can be current or prior. racism If the criminal referred to the victim during, or prior to the shooting, with any racial slur or believed that the victim was inferior or deserving of suffering due to racial reasons THE MIXED METHODS ANALYSIS OF POLICE AND NON-POLICE SHOOTERS 114 Note. This displays the definitions of the variables. Childhood is when the person is of 17 years of age or younger, or if the court states “childhood” as childhood is legally defined as 17 years of age or younger.