University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA VIGILANTISM IN GHANA: EXAMINATION OF PUBLIC ATTITUDES AND MEDIA REPORTS MILLICENT ADZIMAH-ALADE (10598600) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL PSYCHOLOGY DEGREE JULY 2018 University of Ghana http://ugspace.ug.edu.gh STATEMENT OF ORIGINALITY I do hereby declare that this work is the result of my own research and has not been submitted by anyone for any academic award in this or any other university. To the best of my knowledge all references used in the work have been fully acknowledged and due reference is made in the thesis itself. Signed: ……………............. Date: …………………….. ii University of Ghana http://ugspace.ug.edu.gh CERTIFICATION I hereby certify that this thesis was supervised in accordance with procedures laid down by the university. PROF. AKOTIA CHARITY SYLVIA DATE (SUPERVISOR) DR. ANNOR FRANCIS DATE (SUPERVISOR) iii University of Ghana http://ugspace.ug.edu.gh DEDICATION This work is dedicated to all who have suffered any form of vigilantism whether guilty of the crime or not. I bear in your pain and await the day when we can all say goodbye to vigilantism in our society. iv University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENTS It has been a journey of struggle, willpower, persuasiveness and above all, fun. This day wouldn’t have been this joyous as it is without all the people I found myself with. It therefore becomes essential to acknowledge all those who made the journey possible. I would first like to acknowledge God for His mercies and blessings that have brought me this far. It is by His grace that I live and have pursued this programme. I am indebted to my two supervisors for not only guiding me through my work but also serving as my second parent. Prof Akotia Charity Sylvia and Dr. Annor Francis, you folks are one in a million. You nurtured me like your own. Thank God for your gentle souls and the big heart you have to accommodate me and my troubles. You were strong for me, than I was for myself. God richly blesses you. I would also like to acknowledge Prof. Mate-Kole Charles of the University of Ghana Psychology Department. I remember vividly the morning I walked into your office making enquiries about the MPhil programmes in the department and you helped me. Your advice and suggestions have brought me this far. I am grateful, may God richly bless you. Taking time off work to pursue a regular programme was not an easy task. Thanks to my immediate Director and colleagues at work for holding the forte for me so I could pull this through. God bless you all for your help. Special thanks to Mr Aboetaka Selorm, and Mr. Asabere Charles. God bless you and am grateful for the effort and time spent with me. Thank you. A big thank you to all respondents of the survey. Taking time to fill out the questionnaires was time consuming, considering how bulky it was. As such I am grateful to all who took part in the survey. Lastly, to my family the Alades’ and Adzimahs’, you guys are just wow. No words can be used to describe you especially my kids (Erin and Ean) and my better half (Eric), grace has actually found us and we are too blessed, to be stressed. My love to you all. v University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS CERTIFICATION .................................................................................................................................... iii DEDICATION ........................................................................................................................................... iv ACKNOWLEDGEMENTS ....................................................................................................................... v LIST OF TABLES .................................................................................................................................... ix LIST OF FIGURES ................................................................................................................................... x LIST OF ABBREVIATIONS ................................................................................................................... xi ABSTRACT .............................................................................................................................................. xii CHAPTER ONE ......................................................................................................................................... 1 INTRODUCTION ...................................................................................................................................... 1 1.1 Background to the Study ........................................................................................................... 1 1.2 Statement of the Problem .......................................................................................................... 5 1.3 Research Aims and Objectives .................................................................................................. 7 1.4 Significance of the Study ............................................................................................................ 7 1.5 Organization of the Study .......................................................................................................... 8 CHAPTER TWO ..................................................................................................................................... 10 LITERATURE REVIEW ........................................................................................................................ 10 2.0 Overview ................................................................................................................................... 10 2.1 Theoretical Framework ........................................................................................................... 10 2.1.1 The theory of Deindividuation ......................................................................................... 10 2.1.2 The theory of Groupthink ................................................................................................ 12 2.1.3 The Mimetic Theory ......................................................................................................... 13 2.2 Review of related studies .......................................................................................................... 15 2.2.1 Prevalence of vigilantism ................................................................................................. 15 2.2.2 Sources of identity and legitimacy for vigilantism ......................................................... 18 2.2.3 Correlates of attitudes toward vigilantism ..................................................................... 20 Situational correlates ........................................................................................................................ 21 Personal and demographic correlates ............................................................................................... 28 2.2.4 Effects of vigilantism ........................................................................................................ 33 2.3 Summary of literature review ................................................................................................. 34 2.4 Statement of hypothesis ........................................................................................................... 35 2.5 Chapter summary ..................................................................................................................... 36 CHAPTER THREE ................................................................................................................................. 37 vi University of Ghana http://ugspace.ug.edu.gh METHODOLOGY ................................................................................................................................... 37 3.0: Overview ................................................................................................................................... 37 3.1: Study design .............................................................................................................................. 37 3.2 Methods for Study I (Archival study) .................................................................................... 39 3.2.1: Data Sources ...................................................................................................................... 39 3.2.2 Procedure ........................................................................................................................... 40 3.2.3: Data analysis ...................................................................................................................... 41 3.3 Methods for Study II (Survey) ................................................................................................ 42 3.3.1 Population and sampling ................................................................................................... 42 3.3.2 Data collection procedure ................................................................................................. 44 3.3.3 Measures ............................................................................................................................ 45 3.3.4 Pilot study .......................................................................................................................... 47 3.3.5 Data analysis ...................................................................................................................... 48 3.4: Ethical consideration ................................................................................................................ 48 3.5 Chapter summary ..................................................................................................................... 49 CHAPTER FOUR .................................................................................................................................... 50 RESULTS ................................................................................................................................................. 50 4.0 Overview ................................................................................................................................... 50 4.1 Results for Study I- analysis of media reportage on vigilantism .......................................... 50 4.1.1 Socio-demographic characteristics of victims ................................................................... 50 4.1.2 Nature of Vigilantism ........................................................................................................ 55 4.1.3 Forms of Attack ................................................................................................................. 56 4.1.4: Reasons for vigilantism ..................................................................................................... 57 4.2 Results for Study II ................................................................................................................. 63 4.2.1 Preliminary analyses ........................................................................................................ 63 4.2.2 Descriptive statistics and correlations ............................................................................... 72 4.2.3 Hypothesis testing .............................................................................................................. 75 4.2.4 Chapter summary ............................................................................................................. 78 CHAPTER FIVE ...................................................................................................................................... 80 DISCUSSION, RECOMMENDATIONS AND CONCLUSION ......................................................... 80 5.0 Overview ................................................................................................................................... 80 5.1 Discussion of research findings for Study I ............................................................................ 80 5.1.1: Patterns of vigilantism ....................................................................................................... 81 vii University of Ghana http://ugspace.ug.edu.gh 5.1.2: Scope of vigilantism ........................................................................................................... 82 5.1.3: Motivations for vigilantism ............................................................................................... 83 5.2 Discussion of research findings for Study II .......................................................................... 86 5.3: Implications for Theory, Research and Practice ................................................................... 90 5.4: Limitations and Recommendations ........................................................................................ 91 5.5: Conclusion ................................................................................................................................. 92 REFERENCES ......................................................................................................................................... 94 APPENDICES ........................................................................................................................................ 117 Appendix 1: Survey Instrument ........................................................................................................ 117 Appendix 2: Media articles ................................................................................................................ 127 Appendix 3: Ethical approval ........................................................................................................... 135 viii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 3.1 Demographic characteristics of respondents …………………………………... 44 Table 4.1 Reginal distribution …………………………………………………………..… 52 Table 4.2 Victims social demographics ………………………………………………...… 54 Table 4.3 Forms of attack …………………………………………………………...…….. 57 Table 4.4 Factor loadings and eigen values for the various variables being used ……....66-71 Table 4.5 Descriptive statistics of the various variables …………………………….......... 72 Table 4.6 Correlation table………………………………………………………………….74 Table 4.7 Regression model on predictors of attitudes towards vigilantism ……….........…76 ix University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 3.1 Summary of news reports searching process ………………………………...… 41 Figure 4.1 Trends of vigilantism ………………………………………………………… ...51 Figure 4.2 Cross tab of age and extent of damage ………………………………………… 56 x University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS CSPro Census survey processing system MAR Missing at random QCA Qualitative content analysis SPSS Statistical Package for Social Sciences USDS United States Department of State xi University of Ghana http://ugspace.ug.edu.gh ABSTRACT This study examined the phenomenon of vigilantism using a mixed method design, involving an archival study of media reports and a quantitative survey within the Ghanaian context. Drawing on 145 online media reports from January 2002 to December 2017, the study examined trends and patterns of vigilante violence within the Ghanaian context. Results showed a considerable increase in the occurrence of vigilantism within the 15-year period, from 27 cases during the period 2002 – 2011 to 118 cases in 2012 – 2017. Theft and robbery emerged as the most frequently suspected crimes for which victims were attacked. Cases of vigilantism were most frequently reported in urban areas, with the highest number of cases occurring in the Greater Accra Region. In addition, a cross-sectional survey was conducted with 400 individuals selected through convenient sampling strategy to examine public attitudes towards vigilantism. Results from hierarchical linear regression analysis showed that attitudes towards vigilantism were predicted mainly by personal factors (i.e., negative reciprocity belief, extraversion and neuroticism). Situational variables (i.e., police trustworthiness, sense of safety, previous experience of crime, and sense of safety), on the other hand, were not significantly related to attitudes towards vigilantism. The study underscores the implications of vigilantism in terms of disdaining human life and dignity, revisiting of our justice systems, punishment procedures, resourcing and empowering the law enforcers to fight the menace with vigilantism. xii University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.1 Background to the Study On Monday May 29 2017, Captain Maxwell Adam Mahama, a military officer, was lynched at Denkyira Oboase, in the Central Region. According to news reports, he was lynched when residents saw a gun on him during his normal jogging rounds. Owing to an armed robbery incident the night before and suspecting him to be one of the purported robbers, residents raised an alarm which led to the lynching of the officer and the burning of his mortal remains. Also, in May of 2015, two male ex-convicts were lynched and their bodies burnt by some irate youth at Jirapa in the Upper West Region for allegedly stealing motorbikes and ruminants. The youth, fed up with the increasing spate of stealing in the area, and living in the backdrop of several accused persons being released from prison custody by an Appeals Court, went on a rampage meting out instant justice to suspected criminals. Similarly, in April 2011, the sad news of Amina’s predicament went viral on social media when she was arrested by some male students of the Mensah Sarbah Hall Annex B at the University of Ghana, for allegedly stealing a laptop and other gadgets belonging to a female student. The students stripped her naked and physically and sexually abused her and recorded the act. The aforementioned and several related incidents focused considerable national attention on vigilantism and its implication for maintenance of law and order. Haas, Keijser, and Bruinsma, (2012) defined vigilantism as a premeditated criminal act, meted out by a “private citizen” in response to (the supposed or alleged threat of) a crime committed by a “private citizen”, targeting the (alleged) wrongdoer of that crime. Thus, vigilantism encompasses 1 University of Ghana http://ugspace.ug.edu.gh aggressive actions by persons not within the law enforcement profession, which aim to deliver justice in the light of what they consider to be right or wrong; effectively involving taking the law into their own hands (Haas, Keijser, & Bruinsma, 2012). In present times however, vigilantism cuts across religions, traditions, and nations with different names such as: “necklacing”, “lynching”, “mob justice” or “mob action”, ‘instant justice’, and “Xenophobia” (Piper, Charman, & Carciotto, 2016; Tankebe, 2009)1. The act of vigilantism is not a recent phenomenon. The practice originated as a means of deterring and combatting crime. In the late 1700s, public whippings were carried out in Virginia - USA by a vigilance committee led by a Colonel Lynch. The term lynching therefore developed to refer to this practice of public flogging. Over time, the violent punishment meted out by these vigilante bands; and the term ‘lynching’ became synonymous with summary execution, usually by hanging (Tolnay & Beck, 1995). As a means of social control (Tankebe, 2009), vigilantism was also linked with racism in the 19th and 20th centuries, especially in Southern America (Pfeifer, 2004). Not less than 2,500 blacks were killed in the former Confederate states in a time when lynching was rife, translating roughly into about one mob killing every week for five decades (Tolnay & Beck, 1995). These lynching’s which occurred in the 19th and early 20th centuries in the United States, targeted people from particular racial and ethnic groups (Pfeifer, 2004). Fundamentally, these lynchings served to reduce the African American population in these Southern States (Tolnay & Beck, 1995). Campney (2016), reported that in those early years, white newspapers including The Headlight regularly published cases of as well as efforts by whites to lynch and/or terrorize blacks. Every once in a while, these papers reported efforts by 1 For the purpose of this study, vigilantism means same as the other names noted for it in various jurisdictions and will be used interchangeably, most especially with “mob justice” and “lynching” 2 University of Ghana http://ugspace.ug.edu.gh armed blacks to counter those lynchings. Many white indigenes supported the lynching since they felt blacks were responsible for the high crime rates which the legal system was unable to control. Consequently, they saw lynching as a necessary deterrent. In contemporary times, vigilantism was observed world-wide. Harnischfeger (2003) and Minnaar (2001) investigated “necklacing” in South Africa and Nigeria where old car tires were put around the necks of victims and set ablaze for either political reasons, allegations of stealing, witchcraft or other immoral acts. Executions were carried out at the market place and on the street with no statistics to show the total number of persons who were killed (Harnischfeger, 2003; Minnaar, 2001). Most of these victims were killed on the basis of mistaken identity or false accusation (Harnischfeger, 2003; Karmen, 2004; Silke, 2001). Amnesty International (2017) recounted an incident of vigilantism caused by over 1,000 Bengali settlers in Bangladesh. They went on rampage, burned houses of indigenous peoples living in the Chittagong Hill Tracts, killed a 70-year old woman in the process; after they burried a fellow Bengali who was allegedly killed by local indigenous Pahari people. Also, the International Campaign for Human Rights in Iran (2015) indicated that in 2014, several acid attacks were perpetrated against women in the Iranian city of Isfahan for not wearing a hijab, the Islamic headdress. The attacks were undertaken by unidentified men who threw acid in the faces of women who didn’t wear these head dresses that covered their entire faces and heads, leaving only their eyes. Although data is scarce on the phenomenon at the global level, some pockets of data exist on vigilantism in various countries across Africa. Statistics show that the Kenyan police, after including "lynching" for the first time in its crime statistics, recorded 543 cases in 2011. Five hundred and eighty-two (582) people in Uganda also died as a result of lynching in 2014 (Martin, 2010). The story is not different in Ghana. Due to lack of statistical records on 3 University of Ghana http://ugspace.ug.edu.gh vigilantism in Ghana, the specific number of victims who have suffered from the menace cannot be ascertained. However, Adinkrah’s newspaper review in 2005, from 1990 to 2000 recorded 46 incidents of vigilante homicide (Adinkrah, 2005), and cursory review of media reports suggests there has been a rise in vigilantism in Ghana since 2001. A UNODC (2013), report has also indicated a significant rise in violent crimes (vigilantism, theft, e.t.c) in African countries such as South Sudan, Nigeria, Kenya, and South Africa. Vigilantism as a type of social violence comes about when there is a suspicion of wrong doing. Historically vigilantism arose due to issues like theft, heresy, witchcraft, practice of homosexuality, land disputes, political turbulence, religious intolerance, and many others (Adinkrah, 2005; Tankebe, 2009) Usually, vigilantism occurs in two forms; either by individuals spontaneously coming together to mete out justice, or an organized group that goes around indulging in violence in the name of enforcing moral sanity in communities. An initial review of available literature indicates that perpetrators of this act are mainly angry youth between the early 20s to late 30s (Adinkrah, 2005; Tankebe, 2009). Studies both within the Ghanaian context and other nations have grounded the argument on the resort to or support for vigilantism on corruption, circumspection and the slow nature of police and the justice systems (Adinkrah, 2005; Hough, Jackson, Bradford, Myhill, & Quinton, 2010; Sundar, 2010; Tankebe, 2009; Wu, Yuning & Sun, 2009). Thus, those who endorse vigilantism are thought to have low confidence in the criminal justice system due to lack of confidence in the police and the criminal justice system. For example, Tankebe’s (2009), study on vigilante self-help in Ghana found a strong relationship between perception of police trustworthiness and public support for vigilantism. Furthermore, Taussig-Rubbo (2011), asserts vigilantism to be socially recognized, with Ngondi-Houghton (2006), attributing this social 4 University of Ghana http://ugspace.ug.edu.gh acceptance to the consistent failure of the justice system to prevent criminals from coming back onto the streets and to communities from which they were arrested. It becomes abundantly clear then, that though an illegal activity, vigilantism may be seen a necessary response to curbing crime in society. 1.2 Statement of the Problem Vigilantism impinges on the fundamental human rights of people and is effectively against the rule of law. Article 3 of the Universal Declaration of Human Rights states that, "Every person has the right to life, liberty and security of person. “Everyone is entitled in full equality to a fair and public hearing by an independent and impartial tribunal, in the determination of his rights and obligations”. In the Ghanaian constitution, Article 13 of the constitution guarantees the right to life. Clause 1 of Article 13 provides that: “No person shall be deprived of his life intentionally except in the exercise of the execution of a sentence of a court in respect of a criminal offence under the laws of Ghana of which he has been convicted.” Consequently, the act of lynching suspected criminals clearly breaches Article 3 of the Universal Declaration of Human Rights and the Ghanaian constitution (Article 13). In spite of the significant public attention vigilantism received in the country, the subject received little research attention. Most studies on the phenomenon have been done outside Ghana (North & Historical, 2016; Orock, 2014; Outwater, Mgaya & Campbell, 2013; Tankebe & Asif, 2016), making them more Western-centered than Africa-based. The western leanings of these studies may not be applicable to the Ghanaian context as cultures differ, and the reasons and factors that inform the perspectives of people in Western cultures regarding vigilantism may differ from those which pertain in Africa. Veiga, Floyd, and Dechant (2001) argue that a 5 University of Ghana http://ugspace.ug.edu.gh society’s beliefs and values impact how members of that society behave, which may affect how events like vigilantism occur in these societies. Additionally, due to the dominance of criminological research on vigilantism, previous studies on vigilantism emphasized the impact of situational factors like crime rates, procedural justice system and law enforcers (Adinkra, 2005; Outwater, Mgaya & Campbell 2013; Tankebe, 2009). Little is known about the influence of individual factors such as personality and negative reciprocity belief on support for vigilantism. While these studies have contributed to understanding the extent to which normative considerations about the state influences endorsement of vigilantism, they limit our understanding of how individual dispositions shape attitudes towards vigilantism. To advance the knowledge in existing literature, there is the need for psychological studies where the role of individual factors can be assessed. Moreover, empirical research on the phenomenon of vigilantism in Ghana is scarce. Available empirical studies on the subject are those done by Adinkra (2005) and Tankebe (2009). Adinkrah’s study, which was based on the trends and patterns of vigilantism in newspapers, revealed 46 reported cases of vigilantism. It also went further in examining why people indulged in vigilantism. Tankebe’s study, on the other hand, examined Ghanaian vigilantism and the rule of law, with a focus on law enforcers. Although these studies contributed to undertstanding of the nature and correlates of vigilantism in Ghana, additional research is needed for a number of reasons. First, more than a decade after Adinkrah’s media analysis on vigilantism, there were no further studies on trends in vigilantism in Ghana. Yet there is anecdotal evidence that incidents of vigilantism may be increasing in Ghana. Secondly, no study has investigated both situational and personal factors in relation to vigilantism in Ghana. Since conditions necessitating the use of vigilantism might have changed over the years, there is the 6 University of Ghana http://ugspace.ug.edu.gh need for a follow up research to examine the situational and personal factors that impact vigilantism. 1.3 Research Aims and Objectives The study investigated vigilantism in Ghana, by examining public attitudes and media reports. The specific objectives of the study are as follows: 1. To examine the scope, nature, and patterns of Vigilantism in Ghana. 2. To examine the factors that influence the Ghanaian attitudes towards Vigilantism. 1.4 Significance of the Study This study is important for several reasons and to several stakeholders. First, the study contributes to the literature by highlighting how vigilantism dynamics play out in the Ghanaian setting. In effect, it will bridge the cultural gap existing between Africa and the West, regarding vigilantism. Further, the study will close the time gap existing between Adinkra and Tankebe’s study, and vigilantism in recent times. To this effect, the current study will elucidate the trends, nature and scope of vigilantism presently pertaining in Ghana. The study also has relevance for the field of academia especially psychology, as it seeks to examine the role of individual and personal factors in vigilantism. As the study focuses on the Ghanaian attitude towards vigilantism, the findings of the study will highlight behavioral elements which give rise to vigilantism, enabling us justify some psychological theories which pertain to vigilantism. Finally, the study adds to the existing body of knowledge on vigilantism in Ghana and the rest of the world. The findings of the study will lead to a greater appreciation of issues regarding vigilantism and other matters connected with it, thereby stimulating forward-thinking debate and 7 University of Ghana http://ugspace.ug.edu.gh discourse on the subject for enhanced national development. It also has implications for further research and policy in addressing the canker. 1.5 Organization of the Study The study is organized into five (5) chapters. The first chapter gives an introduction to the entire work by providing a background to the study and a context for understanding the work. The research problem being addressed is then presented, in addition to the purpose of the study, the objectives of the study sand research questions. The chapter then proceeds to highlight the significance of the study and discusses the concepts used within which the study is grounded. The scope the study would cover is then delimited, leading to a discussion on the organization of the entire study. Chapter two (2) undertakes a review of theories that can best explain vigilantism and relevant literature pertaining to vigilantism globally, in Africa and in Ghana. Both theoretical and empirical literature is reviewed to give an in-depth grounding to the study and examine attitudes pertaining to vigilantism. In chapter three the methodology and study design adopted in the study is presented. The study area, population, sampling design (sample size and distribution and sampling techniques), data sources and instruments, data management and analyses, ethical considerations and safety protocols adopted in the study are also discussed. Chapter four (4) is the discussion of the findings of the study. Descriptive statistical tests and content analysis were performed on the demographic data of respondents and the media articles on vigilantism, were presented using tables, charts and figures. Following this, respondent’s views and opinions on vigilante violence were analyzed and presented, with inferences and conclusions drawn from the findings. All these findings were subsequently synthesized into a comprehensive report conveying the findings of the study. 8 University of Ghana http://ugspace.ug.edu.gh The entire work was concluded in chapter five (5) through a presentation of the summary, conclusion, and recommendations pertaining to the study. The conclusion highlighted salient parts of the work to aid in better understanding of vigilantism in Ghana. Innovative recommendations were also made relative to the outcomes identified from the study. 9 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW 2.0 Overview This chapter is a comprehensive review of theoretical and empirical literature on vigilantism and other related group-based forms of violence such as mob justice. The theoretical review considered three key theories important to understanding vigilantism. The review of empirical studies was done along three lines; namely the global, African and national levels. Other dimensions covered include prevalence of vigilantism, situational factors impacting vigilantism, sources of identity and legitimacy for vigilantism, support for vigilante violence, and the effects of vigilantism, among others. 2.1 Theoretical Framework This section examined some social psychological theories applicable to the phenomena under discussion. The theories discussed included; theory of deindividuation, theory of group think and the mimetic theory. These theories give insight into vigilantism and together, help us understand how vigilantism evolves and the forms it takes. 2.1.1 The theory of Deindividuation Deindividuation is a social psychological theory which focuses on the loss of self-awareness experienced when people are part of a group. The theory avers that individuals experiencing deindividuation are in a psychological state where their sense of self-evaluation and apprehension evaluation becomes decreased, leading to anti-normative and disinhibited behaviour (Diener, 1976). The theory strives to offer explanations for different types of anti- normative collective behaviour (or group behaviour) such as violent crowds, lynch mobs, and 10 University of Ghana http://ugspace.ug.edu.gh vigilante groups, among other group phenomena (Diener, Fraser, Beaman, & Kelem, 1980). Fundamentally, the theory proposed that when people find themselves in a group, they tend to lose their self-identity and take on the ‘cap’ of the group, thereby engaging in excessive behaviours they would otherwise not have engaged in as individuals. Deindividuation has been generally linked to violent and anti-social behaviour (Silke, 1998). Le Bon first broached the general area of crowd psychology theory and deindividuation with his 1895 publication of “The Crowd: A Study of the Popular Mind”. In this work, Le Bon examined the phenomenon of deindividuation as a function of crowds by examining the concept of crowd mentality, where individual tendencies tend to become dominated by the collective crowd mindset. Le Bon considered crowd behaviour to be “unanimous, emotional, and intellectually weak” (Le Bon, 2017) . He argued for the absence of personal responsibility within individuals in a crowd leads to a tendency for these individuals to behave in a primitive and hedonistic manner collectively as a group. Le Bon further stated that this aggressive or primitive mentality pertains predominantly to the group rather than to individuals within the group, thereby leading to individual traits or tendencies being subdued. Generally, Le Bon was of the view, deindividuation was the state of mind caused by diminished accountability due to greater anonymity as part of a group, whereby the focus shifts from the individual more to the group character and its resulting actions (which in some cases tend to be excessive and extreme) (Silke, 1998). Drawing on Le Bon’s work, Festinger, Pepitone, and Newcomb (1952), then formalized the term deindividuation to describe individuals in a group being treated collectively as a group instead of as specific individuals. They found deindividuation which led to the free display of otherwise inhibited behaviour, was likely to occur when anonymous strangers are together to 11 University of Ghana http://ugspace.ug.edu.gh form a crowd. Festinger et al. argued that when individuals lose their distinctive identities and take on the group identity, they tend to lose control over their internal or moral restraints (Postmes & Spears, 1998). However, Johnson and Downing (1979) were able to show that individuals possess the capacity to connect their self-awareness and peculiar circumstance in inhibiting deindividuation. In light of the above, it becomes easy to see how vigilantism can occur and why people will become quickly involved. Vigilantism makes the group members engage in violent behaviours they would otherwise not have undertaken alone, since an “everyone is doing it” attitude develops within the group. For example, Silke’s study of 1998 on violent assaults in Northern Ireland support the idea that most of the violent aggressions occur with people being disguised or in a mask. To infer from the above, one loses his self-identity or self-awareness when he or she believes he is part of a mob or crowd where he cannot be singled out, thereby making him very likely to engage in vigilantism. 2.1.2 The theory of Groupthink Groupthink was coined by Janis (1972) to explain the phenomenon in which groups make wrong decisions arising from group pressure, which leads to a breakdown of “mental efficiency, reality testing, and moral judgment.” Janis further explained that individuals or groups which suffer from groupthink tend to discount other available alternatives due to pressure to pursue a course of action they have set their minds to, which may sometimes lead them to take irrational actions with dehumanizing and denigrating effects on other persons / groups. It is generally thought that groups are susceptible to suffering from groupthink if they have members with similar backgrounds, if the group is insulated from external influences or opinions, and in cases where no definite or clearly-spelt out rules for decision making exist (Janis 1972, P. 10). Janis (1972) 12 University of Ghana http://ugspace.ug.edu.gh put forth eight symptoms of groupthink, which are: (i) Illusion of invulnerability – which creates excessive optimism that encourages extreme risk-taking, (ii) Collective rationalization – where members discount warnings and do not reconsider their assumptions, (iii) Belief in inherent morality – where members believe in the rightness of their cause and therefore ignore the ethical or moral consequences of their decisions, and (iv) Stereotyped views of out-groups – where negative views of the perceived “enemy” render the effective responses to a conflict seem unnecessary. Others are (v) Direct pressure on dissenters – Members are under pressure not to express arguments against any of the group’s views, (vi) Self-censorship – Doubts and deviations from the perceived group consensus are not expressed, (vii) Illusion of unanimity – The majority view and judgments are assumed to be unanimous, and (viii) Self-appointed ‘mindguards’ – Members protect the group and the leader from information that is problematic or contradictory to the group’s cohesiveness, view, and/or decisions. If these symptoms exist, and groups are highly cohesive and are under extreme pressure to make decisions, groupthink may occur. Janis consistently held that the bond of solidarity that tie people together often causes their mental processes to get stuck. For example, her theory explains the vigilantism among organized groups such as those in Ghana (i.e. the Delta Force, Invincible Forces, Azorka Boys, etc.), which are affiliated to one political party or the other and engage in vandalism and vandalism. 2.1.3 The Mimetic Theory The mimetic theory was propounded by Girard (1977) and holds that if two individuals desire the same thing, rivalry is bound to occur as a result of the desire to possess same. Over time, people forget the object originally desired, after which mimetic conflict transforms the competing interests into general antagonism between or among the parties desiring the object. 13 University of Ghana http://ugspace.ug.edu.gh Instead of pursuing the desired object, people tend to focus on their resentment towards each other, thereby leading to antagonism. This leads to a desire to destroy the other party, who now becomes considered as an enemy. However, an inability to destroy this purported enemy (the other party), leads the parties to seek out a scapegoat on whom to vent their anger. The elimination of this scapegoat through brutal and violent means tends to appease the parties or groups concerned, who then become calm after the violent episode. The scapegoat thus becomes an outlet for deflating pent-up tensions existing in the groups before the occurrence of the violence. Dwelling on the above statements, mimetic theory helps us to explain the rise of violence in democratic and competitive societies. In understanding the conditions in which religion is a cause of international conflict, Thomas (2015) points out that a central part of mimetic theory pertains to the interconnection between culture, religion and violence. Thomas argues that the underlying concepts of mimetic theory such as mimetic desire, mimetic rivalry, the scapegoat mechanism, and veiled and unveiled violence all derive from the origins and maintenance of culture. Instead of arguing for the ambivalence of religion, mimetic theory rather argues for the opposite. It holds violence as the very center of the sacred and argues that violence has always been instrumental in societies’ social conception of culture, religion and the sacred. Focus on the founding myths of several societies shows a linkage with violence, and the role it played in forming a society. This violence is however not visible for people to see every day in society’s cultural institutions. Instead, it is ‘veiled violence’, thus lending credence to mimetic theory’s argument that peace, solidarity and social order come about through violence. A case in point relates to how a particular religion looks down on other religions and employs secular violence 14 University of Ghana http://ugspace.ug.edu.gh against them to avoid confronting bad religious violent practices within its own circles. This is called the scapegoat mechanism. 2.2 Review of related studies Vigilantism comes in different forms and could be perpetrated by already existing groups, or totally unrelated persons teaming up ‘spontaneously’ to achieve a common goal. This makes the phenomenon to be of prime research importance as it can lead to instantaneous, public violence on sometimes unsuspecting victims. This section reviews literature on both forms of vigilantism (already existing groups and spontaneous groups) at the global, continental and Ghanaian level in order to acquire greater insight into the phenomenon. 2.2.1 Prevalence of vigilantism It is difficult to estimate the prevalence of vigilantism. This is especially true when one considers the fact that several cases of vigilantism that occur go unreported. Thus, data on the phenomenon is limited to only those reported cases that make their way into media coverage of the practice as well as police reports on the matter. During historical racial violence that occurred in the late 19th and early 20th centuries, the perpetrators were mainly whites while the victims were mainly ‘blacks’. In those regimes, it was revealed that ‘black’ men, women and children were lynched every six days and suffered not only frequent beatings, whipping, verbal assault but they were also at the mercy of the “white lynch mob” (Tolnay & Beck, 1995). Again, Tolnay and Beck (1995), found that most of the Southern compatriots viewed lynching as a necessary way of justice which guaranteed a swift and severe way of punishing black criminals which required public participation. Available research suggests that countless lives have been lost through vigilantism and the lynchings that took place (Brodie, 2013; Goldbaum, 2014; Tolnay & Beck, 1995). Equal Justice Initiative 15 University of Ghana http://ugspace.ug.edu.gh (2015) reports of 4,075 documented lynching of black people in places such as Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, and Virginia between 1877 and 1950. An estimated 2,805 lynchings occurred between the years of 1882–1930. Most victims of these lynchings were Black (2,462/2,805 = 87.7%). Furthermore, cases where white mobs were identified as lynching blacks constituted the highest incidence of these lynchings (2,314/2,650 = 87.3%) (Tolnay & Beck, 1995). Outwater, Mgaya and Campbell (2013) undertook a study on homicide deaths arising out of mob justice in Dar es Salaam, Tanzania. The study adopted a cross-sectional mixed method design, with surveillance data collected on all 206 victims of “mob justice” in the study area for the year 2005. In-depth interviews were also conducted with persons, including relatives of deceased victims and community members, who survived these violent episodes. The study found that most deceased victims were young adult males, and were identified as being unemployed, thieves, persons unknown in the community, or street vendors. The purpose of the community violence was to defend and protect the community. The violence was more focused towards individuals and could be classified either as lynching or vigilantism; where lynching is an unplanned, spontaneous reaction from the crowd and vigilantism was an organized activity. Outwater et al. (2013) further report that these killings occurred mostly in the streets and public places (75% of deaths), residences/sites of failed burglaries (15%), with the rest (10%) occurring in farmyards/countryside, businesses and in unknown places. The primary weapons used in these deaths were blunt objects (32%), sharp objects (24%), bodily force from fists and feet (18%), fire (20%), and gunshots (2%); with 4% of deaths being caused with unknown weapons. 16 University of Ghana http://ugspace.ug.edu.gh Still in Tanzania, Ng'walali and Kitinya (2006) report that of the 10,000 forensic autopsies undertaken at the Department of Pathology for all reported crime cases in Dar es Salaam, 1,249 were due to mob action. Most of the victims involved were between 15 and 40 years. These age group was seen as unemployed youth who have failed in achieving their dreams and goals. They then result to crime causing increased in crime rates. These deaths were meted out mostly through burning (48.11%) and stoning (49.96%). They therefore conclude mob violence to be a social, legal and public health problem in Tanzania. Ng'walali and Kitinya (2006), further avers that lack of trust in the legal system and corruption, makes the marginalized people resort to mob justice when frustrated. Their observations were further explained as uneven opportunities in rural areas compared to urban areas, increases urbanization thereby leading to increased crime rates in these areas. Adinkrah (2005) examined reports of vigilante homicides in Ghana from 1990 to 2000. The study undertook a profiling of victims of vigilantism murders and the dynamics surrounding their death. Some variables covered in the study include the socio-demographic characteristics of victims, the spatial distribution of these murders, and the circumstances of death. In its execution, the study focused on the scope, nature and patterns of vigilante killings in Ghana, as well as efforts by officials in combatting the phenomenon. The results of the study show that 46 incidents of vigilante homicide were reported from 1990 to 2000. Eighty nine percent (89%) of victims were male compared to 11% who were females. Victims’ ages ranged between 17 and 64 years, with females being mostly elderly women. Victims were mostly of low socioeconomic status, with most of the killings occurring in urban areas. Victims were usually beaten with fists or assaulted with available weapons, sometimes stripped naked, and in certain cases, set ablaze. Reasons for lynching included theft, witchcraft, child abduction, murder, and accusations of 17 University of Ghana http://ugspace.ug.edu.gh genital vanishing by complainants. Three police officers also suffered vigilantism killings in the official line of duty. Some reasons advanced for the prevalence of the phenomenon included heightened public fear of crime, public outrage regarding lenient custodial sentences, the perception that instant justice will deter criminals, lack of training of neighborhood watchdog committee members, strained relations between the public and the police, and the perception of corruption within the police service and courts system. An important weakness of Adinkrah’s study relates to the different timelines from which data was collected. Newspaper reports covered 1990-2000, while interviews with police officers and citizens spanned 2002 to 2004. It is probable that the dynamics surrounding the phenomenon might have changed over the period (between 1990 and 2004). Thus, it becomes problematic reconciling data from both timelines, which effectively raises questions about the validity of the study’s findings due to this mismatch. Furthermore, the author rightly points out, that as with newspaper reportage, the possibility exists for the actual number of occurrences to be higher than that reported. This then means that the prevalence of the phenomenon may be underestimated relative to what actually occurs in society. 2.2.2 Sources of identity and legitimacy for vigilantism Though vigilantism may be interpreted as an agent of state destruction, it sometimes has implicit backing from the state (Harnischfeger, 2003; Minnaar, 2001). Fourchard (2008) relates how vigilantism was a concept introduced by police forces themselves, albeit with positive connotations and intentions in mind. To this extent, it is probable that vigilantism may in certain cases not constitute challenges to state institutions, but actually represent a chosen and preferred alternative type of social organization, instead of being a result of state weakness (Buur, 2008). Alemika and Chukwuma (2005) report that 81% of respondents in their Nigerian study indicated 18 University of Ghana http://ugspace.ug.edu.gh that vigilantism existed in their communities, while 77% intimated that the vigilantes received pay for their services. Aside arising as a response to the seeming ineffectiveness of the criminal justice system, many vigilante groups are also formed in response to the plundering of state resources by political elites (Gore & Pratten, 2003; Ukeje, 2001). Pratten (2008), argued that Nigerian vigilantism is essentially a popular response to criminality, further indicating this vigilantism to be rooted in contexts of embedded rights. South Africa has witnessed an increase in vigilantism groups like the People Against Gangsterism and Drugs (PAGAD) in the Western Cape, which originally rose to fight crime. Over time, they became violent and radicalized (Häefele, 2006). Though considered to occur predominantly in urban areas, some vigilante groups also operate in rural areas as well. An example was the Mungiki vigilante group in Kenya. The Mungiki drew its legitimacy from ethnocentrism, religion, and the anticolonial and intra-ethnic Mau-Mau uprising. They had strong links to political figures who were well-placed in government institutions; though they receive opposition from some state figures and were thus fought by the state based on the party or political figure in power (Anderson, 2002; Kagwanja, 2003, 2005; Wamue, 2001). Some vigilantism groups also subscribed to particular sub-cultures, typified by the vigilante groups of Congo-Brazzaville formed by frustrated youth groups. These youth groups, originally billed as urban defense groups, later turned into armed militias which affiliated themselves with rival political camps and fought in the civil wars that rocked the country in the 90s (Bazenguissa-Ganga, 1999b). The identity of these youth gangs was extensively linked to images of gangs portrayed in the Western media. This characteristic also runs through the Mooryaan vigilante group of Somalia as well as other militant youth vigilantism groups in West Africa (Bayart, 2000; Ellis, 1999; Marchal, 1993; Richards, 1998). 19 University of Ghana http://ugspace.ug.edu.gh Some vigilante groups draw their legitimacy from diverse sources such as traditional, communal, religious and political leaders. Aside their crime fighting functions and connection to political, state and law figures, Nigerian vigilantes also engage in inter-communal violence (Adamu, 2008). Harnischfeger (2003; 2004), reports of how vigilantes in Lagos and southeastern Nigeria engage in not only crime-fighting activities, but target people from other ethnic groups as well. As a case in point, the Bakassi Boys lost their popular legitimacy when they associated themselves with a state governor and were thus seen as a tool for the political elite to use. Violence among various ethnic groups during Nigeria’s Plateau State conflict from 2002 to 2004 also saw vigilantism groups being drawn into the conflict (Higazi, 2008). Higazi reported on how these vigilantism groups shifted from fighting crime during the conflict to offering protection from armed political militias, which then led the vigilantes themselves to more or less become what they were fighting. Atuobi (2008), indicates that dissatisfied African political elements, including leaders and their supporters, sometimes employ violent protests, as well as destroying property, injuring people and killing their political opponents to convey their displeasure with the outcomes of elections. These mostly happens during post-election violence (Lamptey & Salihu, 2012). 2.2.3 Correlates of attitudes toward vigilantism Vigilantism is a social violence that occurs when people are suspected of doing wrong. Some events provoking vigilantism include issues like theft, suspected witchcraft, practice of homosexuality, land disputes, political turbulence, religious beliefs, and many other factors. Most studies conceptualize attitudes towards vigilantism in terms of support for vigilantism. Thus, the extent to which individuals endorse acts of vigilantism. Available literature on correlates of support for vigilantism could be grouped under situational correlates and personal 20 University of Ghana http://ugspace.ug.edu.gh correlates. Studies in Ghana and other countries have located the argument for vigilantism on situational correlates like corruption, circumspection and the slow nature of police and justice systems (Adinkrah, 2005; Tankebe, 2009). Mainly these studies are coming from the criminological literature and could be the reason why individual attributes like the personal factors (personality, belief in just world, negative reciprocity norm and more) were not extensively researched to find their relationship with vigilantism. Situational correlates Tolnay and Beck's (1995) statistical study of lynching in ten southern states in the United States reveals how economic and status concerns are at the heart of vigilantism. Using US Census and historical voting data and a newly constructed inventory of southern lynch victims, Tolnay and Beck identified four reasons why people are lynched. These are for specific deterrence, general deterrence, eliminating or neutralizing African American competitors for social, economic, or political rewards, and as a symbolic appearance of the harmony of white sovereignty. Their findings revealed that blacks living in the American South were more prone to being lynched especially during the period after the Civil War (Wilson, 1978). This is because they believed the blacks were the perpetrators of crime (Tolnay & Beck, 1995). Further, Hagen, Makovi, Bearman and Ware (2013), observed that social change in one form or the other led to the rise of white mobs which lynch black victims. This argument has a historical and sociological basis linking lynching to racial competition or the pressures of social, economic, or political life. This explanation covers historical events like cotton market fluctuations which incited whites (both poor and elite) to develop the urge to dominate black populations by way of lynching (Tolnay & Beck 1995), and citizenship and civil liberties struggles fought by blacks in America (Blalock 1967). Hagen et al. (2013) attempted to create an 21 University of Ghana http://ugspace.ug.edu.gh inventory regarding lynching patterns of lynch mob formation in three Southern U.S. states. Digitized newspaper reports were collated on lynching and near-lynching events in Georgia, Mississippi, and North Carolina between the periods 1882 and 1930; thought to be the pinnacle of the lynching epidemic. The findings show a positive association between prior events which occurred in a county and the likelihood of a completed lynching. Some of these events cover political and economic dimensions. Economically, it was found that a 1-percent increase in cotton dependency led to a 1.47-percent increase in the potential for lynching. Also, greater Republican votes were found to decrease the chances of lynching by 67.8 percent. Ultimately, it was observed that two-thirds of mob formations led to successful incidences of mob kills. In a study by Keil and Vito (2009), on Lynching and the Death Penalty in Kentucky, the researchers through the use of multivariate technique, statistically analyzed black and white lynchings that where done in 1866 – 1934 to find the relationship between legal executions and lynching. It was noted that the Kentuckians took cues from the legal authorities and conceived that if the state executed persons in the name of justice, why shouldn’t they eliminate their enemies? The researchers observed that the lynch mobs simulated the summary executions of convicted criminals (Keil & Vito, 2009). Other reasons given were to prevent riots by blacks, prevent negroes form voting and generally keep the blacks in order as it was believed that the only way to hold black males in check was through lethal force (Greene & Gabbidon, 2000, p.13; Keil & Vito, 2009; Tolnay & Beck 1995, p. 50). Lenta (2014), European study observed that periods following economic and financial crises exerted significant pressure on individuals and societies to adapt to the changing needs of the times they found themselves in. Elena argued that factors such as discrimination, stigma, educational, social and professional exclusions, as well as racist, sexist and nationalist 22 University of Ghana http://ugspace.ug.edu.gh tendencies, all contributed to an environment which sets the tone for various kinds of violence, including vigilantism, to occur as the nation witnessed increasing crime rates. Monaghan (2008), comparing community-based justice in Northern Ireland and South Africa pointed to poor law enforcement as a catalyst for promoting vigilantism. Cook (2013), argues that mob justice arose as a means for people to fight back against the growing crime rates in South Africa, with these crimes mostly being in the form of assault, rape and murder by unscrupulous persons. Minnaar (2001), employed the term ‘vigilantism’ to describe the action of communities using violence in the form of ‘necklacing’ or stoning to death, against people who rob, rape and commit different crimes in the communities. Fanon (1963), argued that mob justice arose due to colonial ideologies being ingrained into the native minds of Africans that led to self-hatred, inferiority, and aggressiveness; which can still be seen in modern times. Fanon further argues that the more violence natives resorted to in order to fight colonialism and secure their independence, the more violence was introduced and accepted into the everyday attitude of natives. It is obvious from the above argument that most violence occurs as a way of communicating one’s displeasure for an action or inaction. Collective crimes are not restricted to the affluent societies or low-income spaces, the aged or the young; it cuts across all demographics, settlement, ethnicity and religions. A key theme in most studies on vigilantism is the public distrust of formal institutions such as the police and other law enforcement institutions which are accused of corruption and improper handling of criminal investigations (Tankebe, 2009; Tankebe & Asif, 2016). Jackson, Asif and Bradford (2014), study have also submitted that vigilantism occurs due to the mistrust, corruption and slow nature of police and justice systems, personal prejudice, hearsay, and a misunderstanding of the criminal justice system (Tankebe, 2009). 23 University of Ghana http://ugspace.ug.edu.gh Ng'walali and Kitinya (2006), argued that peoples’ experience with corruption leads them to engage in mob justice. When people perceive law enforcement as well as judicial officers to be corrupt, they develop the notion that criminals would not be punished for their offences. Consequently, they would prefer meting out their own form of justice, which then leads to situations like vigilante violence and different forms of assault being perpetrated. Tankebe (2010) measured public experience of police corruption at different levels; looking at personal experience (e.g. when people themselves paid bribes to the police), vicarious experience (e.g. witnessing other people pay bribes to the police), and people’s beliefs about how the police leadership is addressing corruption in the police service. Tankebe (2010), found vicarious experience of corruption and the satisfaction people had with the corruption-combatting measures instituted by authorities impacted their assessment of police trustworthiness as well as procedural justice and effectiveness. Furthermore, Tankebe (2009), earlier examined whether perceived procedural injustice contributed to increased public support for violent self-help mechanisms like vigilante justice. In this study, Tankebe (2009) reported public support for vigilante self-help was fundamentally connected to their appraisal of the police. People who perceived the police to be corrupt were therefore more prone to favour vigilantism than those who thought the police were trustworthy. In other words, the lack of trust in law enforcement agencies and the justice system led them to support vigilante activities or take the law into their own hands where necessary. Tankebe and Asif (2016) further investigated the connection between Bottoms-Tankebe model of police legitimacy and support for vigilantism in a survey in Wagah, Ravi, and Gulberg. They found that support for vigilantism had some factors contributing to it; some of which include corruption of the law enforcers, police effectiveness, fairness and lawfulness. The study 24 University of Ghana http://ugspace.ug.edu.gh employed multiple regression analysis in examining the correlates of support for vigilantism. In this study, respondents who had experienced unlawful police behaviour – measured as experience of police corruption, showed greater support for vigilantism. The second finding relates to procedural justice being predictive of support for vigilantism. Specifically, perceived quality of police decision-making was more dominant in influencing perceptions of procedural justice than quality of interpersonal treatment by the police. Thus, the quality of police decision- making was also key in determining whether people would support vigilantism or not. A key weakness of this study relates to the use of attitudinal data, which frustrates efforts at establishing the degree to which respondents’ support for vigilantism can translate into actual involvement in vigilantism, and the circumstances under which it could happen. There is also the need to understand the contextual factors which explain the degree of support for vigilantism across different neighborhoods. Fear and resentment of the police force also contributes to vigilantism (Baloyi, 2015). Tshehla (2014), argues for how governments strive to dissuade people from taking the law into their own hands yet fail to sanction police officers who break the law. In certain cases, policemen (lower ranks) report suffering intimidation from top hierarchy officers, resulting in people losing trust in them (Mathye, 2016). Police officers were sometimes accused of misconduct, unlawful behaviour, brutality and criminality (Baloyi, 2015). Tshehla (2014), reports of close to 9,000 cases of criminal activities committed by South African police officers, leading to doubts about their commitment to fight crime to protect the populace. This further precipitated an increase in vigilantism. Sundar (2010) in a study in India found that the criminal justice system, which was meant to protect the rights of accused persons, and their right to bail, rather negatively impacted their 25 University of Ghana http://ugspace.ug.edu.gh interests in the arena of justice administration. Groupings associated with left-wing organizations like the Communist Party of India (Maoist) and similar Marxist-Leninist-Maoist armed groups (popularly known as Naxalites), constituted another type of vigilantism. These groups held people’s courts in the guise of providing instant justice to villagers who lacked access to formal justice, and in so doing, meted out vigilante violence to accused persons in communities. Furthermore, a women’s group known as the ‘Gulabi Gang’ were known to assault abusive husbands and corrupt officials with sticks and cricket bats in India. These women have been known to hijack trucks transporting food items for distribution to poor people (Prasad, 2008). A recent study by Haas, Keijser, and Bruisma (2012) suggests that situational dynamics impact how people perceive a vigilante act, especially if the crime which led to the ensuing vigilantism is considered to be socially unacceptable or abhorrent. In an experimental study to check for public support of vigilantism with a between group design, Haas et al (2012) revealed that support for vigilantism was not due to lack of confidence in legitimacy of the criminal justice system, but rather the extent or severity of the victim’s crime determine. Example a robber may not be lucky as a petty thief. Various studies also show that communal support for vigilantism self-help, police performance and procedural justice are strongly linked to people’s judgments about the trustworthiness of the police. Procedural justice theory is based on the view that people will obey the law most of the time since they think it is the right thing to do and that compliance is in their own interests (Tyler 2006). A study by Jackson, Asif, and Bradford (2014) on corruption and police legitimacy in Lahore, Pakistan, found that people confer legitimacy, as a consequence, on institutions of justice including the police and the criminal courts. Jackson et al. (2014) also argue that legitimacy is central to procedural justice and may exist among inhabitants when the 26 University of Ghana http://ugspace.ug.edu.gh ruling classes such as law enforcers have authority to command from those they govern and as well as act in ways that validate prevailing power structures. A number of studies have shown that persons who perceived the police as not trustworthy were more likely to support vigilantism (Hough et al., 2010; Jackson et al., 2014; Tankebe, 2013; Tyler, 2004). It has been argued that trust links ordinary citizens to political and legal institutions, thus making the legitimacy of these institutions acceptable to the populace (Adinkrah, 2005; Tankebe, 2009). In addition, Baloyi (2015) argued that the rise in vigilantism is precipitated by the fact that people have lost patience and faith in the criminal justice system, and feel they need to administer justice themselves in order to protect life and property. Jackson et. al (2014) reported several young people as claiming ‘useless police’ as the main reason for vigilantism due to their failure to reduce crime. Tankebe (2009; 2013), corroborated this by stating that anger and resentment are the leading grounds for vigilantism. Sundar (2010), argued the police or security departments of the state also promote a form of vigilantism by virtue of their fight against left- wing groups or movements fighting for self-determination. An example occurred in the Salwa Judum in the central Indian state of Chhattisgarh, where indigenous youth were recruited and armed as special police officers, to fight their fellow villagers who backed the Maoists. This led to the destruction of thousands of homes, the murder of numerous people as well as the rape of women (Sundar, 2010). Ultimately therefore, Sundar argued in distinguishing the various shades of vigilantism (organized violence versus spontaneous acts of mob justice), it would be necessary to consider peoples’ moral views towards the state (and the general state of affairs in general). A key research done within Ghana was that of Tankebe (2009). Tankebe used data collected in 2006 to undertake 2 studies both published in 2009. The first study, which set the tone for the second, concerned an application of the Tylerian model of procedural justice in 27 University of Ghana http://ugspace.ug.edu.gh examining the mediation of procedural fairness in public cooperation with the Police. The second study also employed the Tylerian model to explore emerging phenomenon of vigilante self-help in Ghana. These studies found age, education and police trustworthiness to be the most significant predictors of public support for vigilante self-help. Based on these findings, Tankebe surmised that “the typical profile of a vigilante supporter was an older person with low educational attainment who did not trust the police.” Though mostly true, the finding regarding ‘older’ persons being linked to support of vigilantism did not find support in the literature, as the profile of the average vigilante supporter and perpetrator was portrayed as being that of a male with a youthful characteristic. The key weakness of this study relates to the methodology employed in conducting it. The author hints of the possibility of nonresponse bias, and the limitations associated with cross-sectional data and survey research methodology, as impinging on the findings of the study. To the extent that these weaknesses impact the study, the study’s findings lack wholesale acceptability. Similarly, the sample size of 374 respondents is too small to allow for any meaningful inferences and generalization across the entire population of Accra. Furthermore, general reference to Accra as the location of data collection without specifying the particular enumeration areas data was collected makes it difficult identifying the specific locations the findings pertained to. Thus, factual or empirical verification of the study’s findings cannot be undertaken. Attempts by other researchers to replicate the study in the same, or other areas, would therefore be impossible to achieve. Personal and demographic correlates Unlike situations factors, which have received extensive attention in the literature on vigilantism, little is known about how personal factors are related to attitudes towards vigilantism. An important personal factor considered in this research is personality. Extant research has 28 University of Ghana http://ugspace.ug.edu.gh underscored the influence of personality on aggressive behavior. According to Bushman and Baumeister (1998), certain personality traits can influence aggressive behavior. Some writers have also found a positive correlation between these two variables, showing neuroticism as a strong predictor of anger and hostility (Hofmans, Kuppens, & Allik, 2008; Anderson & Huesmann, 2003; Sharpe & Desai, 2001). In addition, Dukes, Clayton, Jenkins, Miller and Rodgers (2001), study on age and aggressive behavior revealed that age was the most important factor in aggressive driving incidents, with the majority of aggressive drivers being men between 18 – 26 year of age, with personality factors or demographic variables playing a role (Sharkin, 2004). Reiss and Roth (1993), also indicated that aggressive behavior and violence were more in adolescence and increases as they become young adults with increase in injury and death due to weapons being used. These behaviors usually occur in groups or gang activities (lopez & Emmer, 2002) and this most often is done to gain control or power. Mccullough, Bellah, Kilpatrick, and Johnson (1998) conducted a survey in a public university with 91 volunteers from introductory psychology courses where they assessed the relationship between the big 5 personality traits and vengefulness. All participants reported having been offended by another person less than 2 months prior to the study. They assessed the respondents with seven items from Mauger’s Forgiveness of Others Scale. Among the various findings, neuroticism positively correlated to vengeance. The study also revealed that a negative cyclic thinker or someone who likes brooding, is less forgiving of things they have experienced. In addition, Barlett and Anderson, (2012) and Anderson and Bushman, (2002), also underscore a significant relation between agreeableness, extraversion and neuroticism personality dimensions. In ascertaining direct and indirect relations between the Big 5 personality traits and aggressive 29 University of Ghana http://ugspace.ug.edu.gh and violent behavior, Barlett and Anderson (2012), found that extraversion and neuroticism personality dimensions are slightly positive to violence or aggressive behavior. Another important personal variable that may have influence on attitudes toward vigilantism or support for vigilantism is religiosity. According to Cardwell (1980), “religiosity to the psychologist, is addressed in the scopes of devotion, godliness and virtue, whereas sociologists would percieve the concept of religiosity as including church membership, appearance at church, belief acceptance, doctrinal knowledge, and living the conviction”. Bailey and Snedker (2011) examined the relationship between organized religion and lynching in the American South; with a goal to discovering if religious composition influenced lynching. Data for the study came from diverse sources including Beck and Tolnay (2004) who reported 2,805 lynchings in 10 southern states between 1882 and 1930; the Censuses of Religious Bodies (CRB) (1906, 1916, and 1926) which provides membership statistics for American religious organizations by county; the U.S. Census Report on the Statistics of Churches (U.S. Bureau of the Census 1894); and the Inter-University Consortium for Political and Social Research (2001). The study found that counties with greater religious diversity experienced more lynching, showing that larger religious denominations weakened social and moral bonds which kept people from abusing each other, and enhanced white racial solidarity. Sociologists and social historians such as Messner, Baumer, and Rosenfeld, (2006) and Jacobs, Kent, and Carmichael (2005) also studied whether religious contexts support contemporary racism or overlook the use of violence. These were done by assessing the 10 Southern States and a combined explanatory power of both accounts by assessing statistical relations between historical lynchings and the current percentage of African Americans that have suffered lethal conditions through political conditions whiles other plausible effects are held 30 University of Ghana http://ugspace.ug.edu.gh constant. Messner et al., (2006) and Jacobs et al., (2005) believed that a positive relationship exists between the share of local religious adherents who were fundamentalist Christians and the imposition of the death penalty but failed to find a racial effect. Religion was also used as a proxy for the local salience of decency codes, though the attempt to use it in measuring or predicting homicide rates was unsuccessful (Jacobs & Schuetze, 2011). These studies employed a moral solidarity framework to explore the relationship between organized religion and lynching in the American South. Studies like Jacobs et al. (2005) and Jacobs and Schuetze (2011) sought to find out whether a county’s religious composition impacted its frequency of lynching, in addition to the net of demographic and economic controls prevailing in the country. The studies support the notion that a pluralistic religious environment with competing religious denominations weakened the bonds of a cohesive moral community and might have enhanced white racial solidarity and lower incidence of lynching in counties where a larger share of church members belonged to denominations with racially mixed denominations, suggesting that cross- racial solidarity served to reduce racial violence (Bailey & Snedker, 2016; Jacobs et al., 2005; Jacobs & Schuetze, 2011). Aside from personality and religiosity, negative reciprocity belief is another personal variable that may have influence on support for vigilantism. Gouldner (1960) has noted that individuals may endorse a negative norm of reciprocity, under which unfavorable treatment promotes “not the return of benefits but the return of injuries” (p. 172). Indeed, individuals may be guided by negative reciprocity whereby they believe that when someone mistreats them, it is acceptable to retaliate in return (Cropanzano & Mitchell, 2005). Individuals who endorse negative reciprocity belief, believe retribution is the correct and proper response to unfavorable treatment (Eisenberger, Lynch, & Aselage, 2004). Those who hold strong negative reciprocity 31 University of Ghana http://ugspace.ug.edu.gh beliefs are more likely to seek retaliation than avoidance (McLean Parks, 1998). Indeed, research demonstrates that individuals vary in their beliefs about the appropriateness of negative reciprocity. Moreover, individuals’ negative reciprocity beliefs influence behavioral choices (Perugini, Gallucci, Presaghi, & Ercolani, 2003). In sum, individuals with strong negative reciprocity consider retaliation an appropriate response to negative treatment. Negative reciprocity beliefs are rooted in the notion of the just world theory (Lerner, 1966; 1980; Montada & Lerner, 998), which suggests a linkage between what people do and what happens to them (Lerner, 1966) and also that people actually deserve what they get (Montada & Lerner, 1998). Belief in the just world suggests that people who suffer vigilantism, in a way, are partly or fully deserving of the violent experience they suffer. The just world theory would argue that their unfortunate mob encounters could be directly linked to their distasteful behaviors they engaged in. Consequently, proponents of the Just World Theory believe that if actions of persons who suffer acts of vigilantism are examined, it would be found that they are deserving of their fate. In a study building on Lerner’s just world theory, Lodewijkx, Wildschut and Nijstad (2001), conducted an experiment in the Netherlands to determine when violence was deemed senseless. They argued that acts of violence were perceived as senseless when mechanisms by which people maintained the belief in a just world failed to operate. It was observed that respondents perceived violence committed against a victim who was uninvolved with the perpetrator as more senseless than violence committed against a victim who was involved with the perpetrator. The presence or absence of negative victim attributes did not significantly affect participants’ perceptions of senselessness. Furthermore, there is the outstanding situation where violence may be perpetrated when one feels the victim deserves to 32 University of Ghana http://ugspace.ug.edu.gh be punished. This can generate into negative reciprocity, thus stressing the biblical order of “a life for a life, eye for eye, tooth for tooth . . . bruise for bruise” (Exodus 21:23–25). 2.2.4 Effects of vigilantism The 19th and 20th century incidences of lynching caused much trauma amongst black people in the U.S. with little opposition from state and federal officials. Akin to race-based terrorism, these “terror lynching” reached extreme proportions between 1880 and 1940; causing great fear and humiliation within African American communities. The net effect of the terror lynching was "mass migration by millions of blacks from the South into urban ghettos in the North and West throughout the first half of the twentieth century”, and the creation of a culture of racial subordination and segregation for several decades (Equal Justice Initiative, 2015). Hamber and Lewis (1997) indicate that the mob way of dispensing justice seems quicker and more effective, and therefore restores peoples’ faith in the justice system. Vigilantism however, has implications, most of which are usually negative. Baloyi (2015) argues that children and some adults who watch these kinds of gruesome killings are exposed to not only blood and death, but also to the violence portrayed during the lynching. Consequently, such impressionable ideas are picked by children and adults who witnesses these violent acts. They might think vigilantism is the right way of dispensing justice in society, thereby perpetrating it themselves. The net effect of this is an inculcation of a lack of respect for human life and dignity into the next generation of citizens, who will have no qualms about dragging suspected criminals through the streets, beating them, and finally burning them to death. Stavrou (2011) presents some of the effects of violence, of which mob violence is a type, as including: Emotional changes characterized by numbing, powerlessness, extreme vulnerability and a feeling of lack of safety. Additionally, anxiety, irritability and restlessness also become 33 University of Ghana http://ugspace.ug.edu.gh part of this result. As well, difficulty in sleeping resulting from nightmares about the attacks and Social isolation in which children will normally not want to socialize and will usually be aggressive. These effects of violence are all evidence of the existence of post-traumatic stress disorder (PTSD) and need to be treated as such. Vigilantism also tends to trivialize the value of human life and dignity. When a person is beaten, stabbed, maimed or burnt over the accusation of stealing a mobile phone, chicken or money, the mob gives the impression that the value of the human life being taken is equivalent to the value of the phone, chicken or money stolen. A key tenet of penology, the science of punishment, is that punishment meted out should not exceed the gravity of the offence committed. Thus, when a person is killed or burnt for allegedly stealing a mobile phone and society accepts such killing, what both the mob and society are effectively saying is that the value and worth of human life is equivalent to the price of a phone. In most cases however, phones are cheap, and human life is priceless. The value and dignity attached to human life should be enough to cause mobs to exercise restraint in meting out their form of justice (Adu- Gyamfi, 2014). Thus, vigilantism has latent and indirect effects which can push the incidence of violence in society to higher proportions through the perpetrating of reprisal attacks (Baloyi, 2015). 2.3 Summary of literature review A key theme in most studies on mob justice is the public distrust of formal institutions such as the police and other law enforcement institutions which are accused of corruption and improper handling of criminal investigations. Most studies have also argued mob justice as occurring due to the mistrust and slow nature of police and justice systems, personal prejudice, hearsay, and a 34 University of Ghana http://ugspace.ug.edu.gh misunderstanding of the criminal justice system. (Jackson, Asif, & Bradford, 2014; Tankebe, 2009; Tankebe & Asif, 2016). Aside the situational factors, some personal factors like negative reciprocity belief is seen to have direct relationship with violence (Eisenberger, Lynch, & Aselage, 2004). As well study by Barlett and Anderson (2012), also found some positive relationship between extraversion, neuroticism and agreeableness personality dimensions. In effect, vigilantism can be seen to be a consequence of both personal factors as well as situational factors such as mistrust of justice delivery systems. 2.4 Statement of hypothesis Hypothesis 1: Demographic variables will be significantly related to attitudes towards vigilantism. Specifically: 1a) Age will have a significant relationship with attitudes towards vigilantism (support for vigilantism). 1b) Sex will have a significant relationship with attitudes towards vigilantism. Hypothesis 2: Situational factors will be significantly related to attitudes towards vigilantism. Specifically: 2a) Police trustworthiness will have a significant relationship with attitudes towards vigilantism 2b) Sense of safety will have a positive relationship with attitudes towards vigilantism 2c) Experienced of crime will have a significant relationship with attitudes towards vigilantism 35 University of Ghana http://ugspace.ug.edu.gh Hypothesis 3: Personal factors will be significantly related to attitudes towards vigilantism. Specifically: 3a) Negative reciprocity beliefs will positively relate with attitudes towards vigilantism. 3b) Personality dimensions (extraversion and neuroticism) will positively relate to attitudes towards vigilantism. 3c) Religiosity will have a negative relationship with attitudes towards vigilantism. 2.5 Chapter summary This review goes to show that vigilantism is a topical issue that cut across the globe and is eating deep into the contemporary society as more people are resorting to this form of justice at the expense of the main procedural justice system. Enhanced efforts should therefore be pursued in curbing vigilantism issues in countries worldwide. The next chapter takes us to the methods and procedures used in conducting the research. Much is discussed about how data was retrieved and respondents sampled. 36 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE METHODOLOGY 3.0: Overview This chapter discusses the research design, population, sampling techniques and sample size, sources of data collection, and data collection instruments employed in the study. It also discusses the data management and analysis procedures adopted in the study and the ethical consideration procedures employed when dealing with respondents in the study. The chapter is divided in two sections, that is study I and study II. A general study design is provided for the two studies, there after the other presentations are done separately for the various studies and concluded with ethical consideration. 3.1: Study design The study adopted a concurrent mixed methods design involving two phases, a qualitative phase (Study I) and a quantitative phase (Study II). According to Creswell, Clark, Gutmann, and Hanson (2003), the concurrent mixed method design is an amalgamation of two or more methods employed in confirming, cross-validating, or substantiating results of a research, whereby data collection is simultaneous. Because mixed methods research permits for possible triangulation to examine the same phenomenon, it allows the researcher to identify aspects of a phenomenon more accurately by approaching it from different vantage points using different methods and techniques. By this argument, more than one method was employed in order to minimize the weaknesses inherent in using each method and to maximize their strengths. Study I adopted an archival research design in which data was collected from online published articles of some credible media houses in Ghana. Archival study is a type of primary 37 University of Ghana http://ugspace.ug.edu.gh research that involves seeking out and extracting evidence from primary sources held in an archive, a special collections library, or other repository (Pearce-Moses & Baty, 2005). Archival sources can be manuscripts, documents, records (including electronic records), objects, sound and audiovisual materials, or other materials. The main objective of the archival study was to examine the scope, nature and patterns of vigilantism in Ghana. Since vigilantism is a topical issue and the researcher sought to understand the scope, nature and patterns of vigilantism, the archival design was selected because it would enable the researcher to study behavior that develops over decades (Mullen, Kharbanda, Cross, Donald, Taylor, Vallance & MacAllister, 2001). Online media reports screened and included in the final data set of vigilantisms were those reporting instant justice, mob action or vigilantism of any kind in Ghana. Study II employed a survey which was cross-sectional in nature using the public specifically people within the Greater Accra region. Cross-sectional survey is a research tool used to capture information based on data gathered for a specific point in time. Time constraint was a major consideration in the adoption of a cross-sectional design. Cross-sectional methods are relatively quick and easy to conduct, allow data to be collected once on all variables and also able to measure prevalence for all factors under investigation. Although cross-sectional designs have the inherent limitation of precluding causal inferences, they provide a useful means to gathering descriptive data on social psychological phenomena. The remainder of this chapter presents the specific methodological steps and analytical approaches adopted in both Study I and Study II. The chapter ends by explaining how ethical issues were addressed in the study. 38 University of Ghana http://ugspace.ug.edu.gh 3.2 Methods for Study I (Archival study) 3.2.1: Data Sources The method for study I was modelled after the media (news) surveillance method used by recent studies by Quarshie and his coleagues (Quarshie, Osafo & Akotia,2015; Quarshie, Osafo, Akotia, Peprah & Andoh-Arthur, 2017) on some social and legal issues in Ghana. Data was drawn from media sources, specifically the websites of newspapers, radio stations, some news agencies, and digital (online) media houses. The newspapers used are Daily Graphic, Daily Guide, and The Ghanaian Times. The radio stations are Joy FM (www.joyfmonline.com), Citi FM (www.citifmonline.com), Adom FM (www.adomfmonline.com) and Peace FM (www.peacefmonline.com). The general news agencies are Ghanaweb and Ghana News Agency. The selected newspapers (Daily Graphic, Daily Guide, and the Ghanaian Times) are the key newspapers in Ghana (Prah & Yeboah, 2011). Daily Graphic is a state-owned daily newspaper, and is the most widely circulated newspaper in Ghana (Gadzekpo, 2010). The Ghanaian Times is another state-owned newspaper which reports authoritatively on social and political issues, with Daly Guide being a private newspaper. Joy FM (myjoyonline.com) and Peace FM (peacefmonline.com) are the most listened to radio stations in Ghana (Adinkrah, 2014; Prah, & Yeboah, 2011). Ghana web (www.ghanaweb.com) and the Ghana News Agency (ghananewsagency.org) provide detailed stories about topical news of issues including health; business and economics; politics; science; education; and sports. These media sites were used because of their vast coverage areas and accessibility within the Ghanaian society. 39 University of Ghana http://ugspace.ug.edu.gh 3.2.2 Procedure The study focused on all articles that have element of vigilante violence within the Ghanaian context within the period of 2002 to 2017. In the present study, there was no initial decision made for how many articles to be used for the study. Rather, all published articles on vigilante violence in Ghana published by the selected media sources within the period under study were selected for review. To guide the extraction of these media articles, Haas, Keijser, and Bruinsma,’s (2012) definition of vigilantism was used. Other various names known to reflect vigilantism, example “necklacing”, “mob justice” and “instant justice”, by other writers were also used (Lenta, 2014; Piper, Charman, & Carciotto, 2016; Tankebe, 2009). Duplicated stories or multiple account stories were treated as one. That is if there are 5 duplicates of a story, it will be treated as 1 instead of 5 separate ones. An archival search of the website of each of these selected media sources was done using keywords such as ‘‘mob justice,’’ ‘‘vigilante violence,’’ ‘‘mob action,’’ ‘‘instant justice,’’ “witch”, “rape”, “thief”, “land dispute”, “election turbulence”, “arm robbery or robber”, “political misunderstanding”, “burnt alive”, “beaten to death”, “vigilantism”, “street justice” and ‘‘lynching.’’ This allowed for incidence of vigilante violence to be captured for better analysis. Online media reports screened and included in the final set of vigilante violence were those reporting instant justice or vigilante violence of any kind in Ghana. To guide the extraction of these media articles, Haas, Keijser, and Bruinsma, (2012) definition of vigilantism was used. Other various names known to reflect vigilante violence by other writers were also used (Elena, 2014; Piper, Charman, & Carciotto, 2016; Tankebe, 2009). However, all international media articles on vigilante violence before and within the stipulated frame of study were excluded. Duplicated stories or multiple account stories were treated as one. That is if there were five 40 University of Ghana http://ugspace.ug.edu.gh different reports on the same story, it would be treated as one instead of five. A summary of the steps in selecting and screening media reports is depicted Figure 3.1 Summary of news reports searching process 3.2.3: Data analysis Media reports on vigilante violence was drawn and subjected to qualitative content analysis- QCA (French, Anderson, Burnard, Holmes, Mashaba, Wong, & Bing- hua, 1996; Graneheim & Lundman, 2004; Quarshie et al., 2015, 2017; Schreier, 2012) to determine the main trend and pattern of vigilante violence in terms of prevalence, sociodemographic characteristics, causes, risk factors, and methods used. Frequencies were also run for demographical data as well to check for trends over the years. The QCA proceeded in the four-stage method proposed by French, Anderson, Burnard, Holmes, Mashaba, Wong, and Bing- hua (1996). At the first stage, all the cases of vigilantism that were drawn from the selected media sites were read and re-read in order to be conversant with the reports and by hand, note as many credible or reasonable initial ‘‘open codes’’ as possible across the entire data set that will be generated. In the second 41 University of Ghana http://ugspace.ug.edu.gh stage, the focus was on reading the articles more closely and approving on a set of initial codes which were considered very relevant to the data in order to reduce the number of words and phrases so as to produce a manageable list of headings that account for all the data in the materials. Thirdly, there was an integration of all relevant generated initial open codes that were similar to initial codes which aided the search for categories and emerging themes that help explain larger sections of the data. Finally, the last stage of the analysis, data was reviewed and refined of the themes and selected extracts, which support and described the themes. Furthermore, meanings were given to emerge themes and categories in the light of theory, previous related studies, and the general Ghanaian sociocultural context. To improve validity of the analysis process, codes and themes were thoroughly discussed and agreed upon by researcher and supervisors before further analyses was done. Cross-validation and group interpretations were done to reduce bias and increase the credibility and trustworthiness of the research findings (Whittemore, Chase, & Mandle, 2001). 3.3 Methods for Study II (Survey) This section describes the methods used in the quantitative phase of the study. Specifically, section also presents sampling methods, measures, and procedures employed in data collection as well the analytical strategies used in the quantitative phase of the study. 3.3.1 Population and sample The population of a study encompasses all persons who possess characteristics desirable in respondents. It pertains to the aggregate or totality of people, objects or members who conform to particular specifications sought in a study (Polit & Hungler, 1999). In the current study, the 42 University of Ghana http://ugspace.ug.edu.gh targeted population was people living in Accra, within the Madina, Adenta, Haatso and Legon enclaves. Consideration was given to people of sound mind who can read and write without any difficulties. The study also considers the general Ghanaian population so as to get accurate data that can be generalized. There was a consideration for age in determining who qualifies to partake in the study. Only those within the legal age of 18 years and above, were allowed to partake in the survey. Further, the survey made use of the general public who can read and write to avoid the limitations that comes with mistranslation of questionnaires and being able to finish within the stipulated time given for completion of work. Focusing on people within Accra also helped in accessing divergent views from different ethnic background since these are different people from all parts the country. The survey included all categories of people with different occupational background who could read and write. A non-probability sampling technique, specifically convenience sampling was employed for the study. Convenience sampling enabled the researcher have easy access to the number of respondents needed for the study. In all 400 individuals participated in the study. These respondents were drawn from the general population of Accra. The expression by Krejcie and Morgan (1970) in determining sample size from an unknown population was adopted in determining the study’s sample since the entire population was unknown to the researcher. Forty- nine percent of respondents were aged 21 to 30 whilst the least respondent group was 50 years and above. Age categorizations were done base on the following demarcations; teenagers (18 - 20), adolescents (21 – 30), young adults (31 – 40), adults (41 – 50) and the aged (50 years above). Further, male respondents made up 65% of the survey whilst the females made up the rest 35%. 90% of respondents were Christians and 10% were Muslim. Respondents occupation were also categorized into Student representing 27%, Public / Civil service 45%, Self-employed 43 University of Ghana http://ugspace.ug.edu.gh 16%, Unemployed 2%, Artisans 3% and No response 7%. The demographic dynamics of the study’s respondents is presented below. Table 3.1 Demographic characteristics of respondents Item Categories Frequencies Percentages (%) 18-20 years 64 16% 21-30 years 196 49% Age 31-40 years 108 27% 41-50 years 16 4% 50 years + 16 4% Sex Male 260 65% Female 140 35% 120 Marital Status Married 280 30% Single 70% Christian 360 90% Religion Muslim 40 10% Student 108 27% Public / Civil service 180 45% Self employed 64 16% Occupation Status Unemployed 8 2% Artisans 12 3% No response 28 7% Source: Field Data, 2017 3.3.2 Data collection procedure After approval was given by the Ethics Committee for Humanities (ECH - University of Ghana), the researcher set out to conduct the survey (ECH 054/17-18). In every area or location that data was collected, proper introduction of both the study and researcher was done. Respondent consent was sought prior to data collection, with the purpose of the study explained to them as well. Clarification was then made to respondent on issue regarding the questionnaire that they might have and if not, the questionnaires were given to them after which they were asked to read 44 University of Ghana http://ugspace.ug.edu.gh carefully before given their consent. At the end, all completed questionnaires were placed in an envelope provided. This was to ensure confidentiality of participants’ responses. 3.3.3 Measures The materials for the survey compiled into one questionnaire with different sections measuring various variables. The questionnaires were adopted from previous related studies (Haas, Keijser, & Bruinsma, 2012; Sunshine & Tyler, 2003; Tankebe, 2009) which was measured on a five- point Likert scale and modified to suit the current study and setting. The questionnaire measured eight different variables, thus general demographic data, previous experience of crime, negative reciprocity belief, attitude towards vigilantism, perception of police trustworthiness, sense of safety, two personality types noted to be prone to aggressive behaviours, and person (s) religiosity. The portion of the scale that measured previous experience of crime, asked respondents to give a yes or no rating to the questions as follows: Have you ever been a victim of any kind of violence (Theft / Robbery, Political turbulence, Land dispute Rape and Mob justice)? Have you witnessed any of these forms of violence before? Where yes was coded as 1 and no as 2. Cronbach alpha could not be computed for experienced of crime in the present study. The negative reciprocity beliefs scale by Eisenberger, Lynch and Aselage (2004), was adopted, and used to assess the people’s belief in negative reciprocity. The scale has an alpha of α=.85 which made it reliable and suitable scale for the study. The scale had 15 items, which was rated on a five-point scale ranging from1 = strongly disagree to 5 = strongly agree. The following questions were asked; I belief in the saying, “A tooth for a tooth and an eye for an eye”. Anyone who treats me like an enemy, deserves my resentment. The present study gave Cronbach α =.86. 45 University of Ghana http://ugspace.ug.edu.gh Attitudes towards vigilantism was measured using support for vigilantism scale by Sunshine and Tyler (2003), with an alpha of α=.86. This was also modified and questions framed to reflect the Ghanaian environment. The questions were; (1) It is alright for members of the public to beat up crime suspects, (2) People who kill armed robbers should not be blamed, (3) I don’t belief in the right of a criminal and more. The items were rated on a five-point scale. Cronbach alpha was same as the originators own (α =.86). Perception of police trustworthiness scale adopted from Tankebe (2009) and Haas et. al, (2012), indicating an alpha of α=.96 was used in measuring police trustworthiness. Respondents were asked to rate the extent to which they consider the Ghana Police to be trustworthy on a five- point scale. Responses were rated on a five-point scale ranging from ‘strongly agree’ to ‘strongly disagree’.” Sample items on this scale included; the Ghana police are trustworthy and the police can be trusted to make decisions that are right for the people in my neighbourhood. The Cronbach alpha reliability coefficient for this measure in the present study is α=.80. Another key variable tested for was sense of safety. Respondents were to rate the question asked on a five-point Likert scale where 1=Strongly Disagree, 2=Disagree, 3=Neutral, 4=Agree and 5=Strongly Agree. Overall, my neighbourhood is a good place to live in terms of security. Crime levels in my neighbourhood have changed for the better in the last year. The last question which was; To what extent do you feel safe in your neighbourhood? Was rated separately using unsafe to very safe. This was done using the sense of safety scale by Migliorini, Rania, Cardinali, and Manetti (2008), with α=.75. The present study indicated an α =.65. Inferring from other studies, it was noted that Extraversion and Neuroticism personality types are prone to aggressive behaviours. To understand which of the two-personality types influence vigilantism behaviour, the respondents were asked to rate the following question on a 46 University of Ghana http://ugspace.ug.edu.gh five-point Likert scale. Personality scale developed and used by Costa and McCrae (1992), with a Cronbach alpha reliability coefficient of α=.78, was used in measuring the two personality types to attitudes towards vigilantism. Example of questions that measured for extraversion are I am a talkative person and I like plenty of movement and excitement around me. Neuroticism was measured using questions like, my mood often goes up and down and I am troubled about feelings of guilt. In addition, Hendricks (2006), religious well-being scale with α=.94 adopted in measuring people’s religiosity to know if it will predict attitudes towards vigilantism. Respondents rated their spirituality levels on a five-point scale ranging from Every day =1; Most days =2; Some days =3; Once in a while =4; and Never or almost never =5. Respondents were asked to rate their religiosity on questions such as; (1) I feel God’s presence and (2) during worship, or at other times when connecting with God, I feel joy which lifts me out of my daily concerns. An α =.84 was obtained in the present study. 3.3.4 Pilot study Before the actual data was collected, a pilot study was conducted within the Department of Psychology of the University of Ghana. For this study, a sample size of 15% of the projected total sample size was used in the pilot study (Connelly, 2008; Lackey & Wingate, 1997). According to Hertzog (2008), “samples as small as 10% -15% per group sometimes are sufficient” for a feasibility study (p. 190). Hence, the used of 15% of the actual sample size for the pilot study. This resulted in 50 out of 400 under graduate students being used. convenience sampling was used in piloting because it allows the researcher to obtain basic data and trends regarding the study without the complications of using a randomized sample. The purpose for the pilot study was to evaluate the items on the scale (measures) used and also to seek feedback on 47 University of Ghana http://ugspace.ug.edu.gh the measures. This helped reframe and make changes to some phrases that were used in the questionnaire. It also helped in clearing some ambiguities that could limit the study in the future. 3.3.5 Data analysis Data was first keyed into census survey processing system (CSPro) and errors or data inconsistencies checked and corrected before exporting to Statistical Package for Social Sciences (SPSS) for further analyses to be performed on the quantitative data collected. A hierarchical linear regression was also performed to find the associations between the variables under investigation and attitude towards vigilantism. Those Variables investigated include; age, Perception of police trustworthiness, religiosity, the two personality types, sense of safety, support for vigilantism, experience of crime and negative reciprocity belief. Respondents’ assessment of attitudes towards vigilantism in respect to support for vigilantism is presented in percentages. Various ages were grouped into; teenagers (18 – 19), adolescents (20 – 29), young adults (30 – 39), adults (40 – 49) and 50 years above forming the aged respectively. 3.4: Ethical consideration Ethical protocols were observed by the researcher in dealing with respondents and the consent of respondents was sought for before data was collected from respondent. First, the purpose and objectives of the study was explained to respondents. This way, respondents know what the study is about and get a better understanding of It and they are free to withdraw from the research as and when they think they can’t partake in it anymore. Also, the researcher explained to respondents how their demographic data would be handled. This information will not be disclosed in any way that it can be traced to the respondent. Should the need be to disclosed such information, the respondent will be traced for consent before the researcher will go ahead. 48 University of Ghana http://ugspace.ug.edu.gh Fundamentally, the researcher assured respondents of their safety, privacy and confidentiality; and the fact that their personal information’s in the form of demographics were used for academic purposes only. Observing these ethical measures ensued to the safety and security of respondents not in any way compromised, in addition to the protection of their identities and privacy. No personally identifiable data was therefore collected from respondents as in their names. The researcher in light of the topic understudied will refer for a psychologist for care and treatment but done base on consent from respondent to help anyone who will have any psychological issues arising from the study to ensure no harm comes to respondents from participating in the study. 3.5 Chapter summary The chapter outlined the various ways by which the study was conducted and the various analysis done. It started with the study design, then the data source, population and sampling, measures and concluded with ethical consideration. The next chapter leads us to results presentations from both studies. It is also divided in two sections, Study I and Study II same as how this present chapter was put across. Results for Study I presented first, followed by result for Study II. 49 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR RESULTS 4.0 Overview The chapter presents the results of both analysis done in the archival study and the survey conducted on attitudes towards vigilantism. The chapter is divided into two sections. Results of study I are presented in the first section and the subsequent section contains the result of study II. The chapter is then concluded with a summary of the entire chapter. 4.1 Results for Study I- analysis of media reportage on vigilantism This section presents the analysis of data collected from online news sources on vigilantism. The socio-demographic characteristics of victims and trends of vigilantism is presented first, followed by the nature of vigilantism, then the motivating factors that give rise to vigilantism in Ghana. Themes deduce from the cases as being reason for vigilantism are presented lastly to throw more light on vigilantism in Ghana from the media perspective. 4.1.1 Socio-demographic characteristics of victims There was generally an upward trend was observed from 2002 to 2017 (Figure 1). A consistent low incidence of vigilantism reportage was recorded between 2002 and 2006. From 2007, however, a slight increase was recorded till 2009 when it dipped. From that point onwards, the incidence of vigilantism increased all the way to 2015 (representing 16.6% of all cases). A fall in incidence was recorded in 2016 (representing 13.1%). The year 2017 saw 42 cases reported, which represents 29% of all reported cases. Majority of the cases (87.7%) were reported from 2011 to 2017; even that, 29% of the cases were reported in 2017 alone. From 2002 to 2006, single (1) cases of vigilantism were reported respectively. This could mean a non-occurrences of 50 University of Ghana http://ugspace.ug.edu.gh vigilantism cases or virtually the cases were not reported in media. In 2007, there were 4 reported cases, 2008 witnessed 5 cases, 2009 had 2, 2010 were 3 cases, with 2011 and 2012 having 9 cases respectively. In the year 2013, there were 10 cases, 2014 had 14 cases, 2015 witnessed 24 cases, 2016 also witnessed 19 cases and in 2017 41 cases were reported being the chunk of the cases retrieved from online. This could reflect that more of the cases of vigilantism are being reported in the media than previously. finding their way into the media than the previous years. 35.0 29.0 30.0 25.0 20.0 16.6 15.0 13.1 9.7 10.0 6.2 6.2 6.9 5.0 2.8 3.4 2.1 .7 .7 .7 .7 1.4 .0 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 YEAR Figure 4.1 Trends of media reports on vigilantism in Ghana from 2002 to 2017 Source: Field Data, 2017 Cases of vigilantism were seen to have occurred most in the urban areas, with Greater Accra Region accounting for the highest 29.7% (43 reported cases). This was followed by the Ashanti Region accounting for 28 cases representing 19.3%, and Upper west region with 11.7% (17 51 PERCENTAGE OF OCCURANCE University of Ghana http://ugspace.ug.edu.gh reported cases). Volta and Eastern regions accounted for some of the least reported cases with 1 and 3 cases representing .7% and 2% respectively. Broadly, the results further show that majority of vigilantism (74%) occurred within urban areas compared to the rural settings (24%). Two percent (2%) of violence locations were unreported. Table 4.1 Regional distribution of cases of vigilantism REGINAL DISTRIBUTION FREQUENCY PERCENTAGE (%) Greater Accra region 43 29.7% Ashanti Region 28 19.3% Upper West Region 17 11.7% Upper East Region 14 9.7% Northern Region 12 8.3% Central Region 10 7% Western Region 9 6.2% Brong Ahafo Region 6 4% Eastern Region 3 2% Volta Region 1 .7% Not Stated 2 1.4% Source: Field Data, 2017 Victims of vigilantism are from different age groups. The youngest were aged between 10 - 19 years (9.7%), followed by those within the age group of 20-29 (22.1%). After these came victims aged 30 - 39 years (15.2%), followed by those aged 40 – 49 years (3.4%) and 50+ years (4.1%). The ages of about 45.5% of victims were not captured in the media reports. Age groupings were done to reflect teenagers, adolescents, young adults, adults and the aged respectively. It is also 52 University of Ghana http://ugspace.ug.edu.gh evident from the table that more males than females experience vigilantism in Ghana. For example, 89% of victims were males compared to 8.3% who were female. In instances of political turbulence, it is not clear whether it was males or females who suffered the act of vigilantism. Thus, they representing 3% of victims in the gender analysis. Victims of vigilantism in this study were found to have some form of employment. Though the professions of 69% of victims were unreported, 13% were employed, 12% were self- employed and 6% were unemployed. Those in formal employment occupied different positions including Hospital Administrator, Soldier, Member of Parliament, Police, Preacher man, constituency chairman and a security coordinator. Self-employed victims included traders, famers, mechanics, drivers, truck pushers, galamsey operators, and a fetish priest. Students were also not exempted from experiencing vigilantism, as victims of 3 cases reported were student but was captured under unemployed for the purpose of analysis. The various analysis of socio- demographic data is captured in Table 4.2 53 University of Ghana http://ugspace.ug.edu.gh Table 4.2 Victims social demographics Social demographics Frequency Percentage (%) Age group Not reported 66 45.5% 10-19 14 9.7% 20-29 32 22.1% 30-39 22 15.2% 40-49 5 3.4% 50+ 6 4.1% Gender of Victims Male 129 89% Female 12 8% Not reported 4 3% Occupation of Victims Employed 18 12.4% Self employed 18 12.4% Unemployed 9 6.3% Not reported 100 68.9% Location of Incidence Urban 107 74% Rural 35 24% Not reported 3 2% Source: Field Data, 2017 54 University of Ghana http://ugspace.ug.edu.gh 4.1.2 Nature of Vigilantism The results indicated generally that 98 victims, representing 67.6% of the 145 cases of vigilantism recorded died through the incident 47 victims (32.4%) survived the violence. Figure 4.1.2 indicates that for 66 out of the 145 victims whose ages were unreported, 44 of them lost their lives through the violence meted out to them while 22 survived. Likewise, victims within the ages of 10 – 19 had majority surviving (12) with only 2 dying. It was further observed that victims above 40 years did not survive the attacks Also, 8 of the victims between the age range of 20 – 29, survived the violence while 24 people died after experiencing vigilantism. Similarly, for the age range of 30 – 39, 5 victims survived while 17 victims died. It was observed that the age group that seems to have suffer most damage in terms of death is the age between 20 to 29. It was obvious that those whose ages were not captured in the cases reviewed comparatively recorded the highest of death but people within this category could fall under any of the age groupings done. The age group of 10 to 19 indicated the least suffered in terms of death. 55 University of Ghana http://ugspace.ug.edu.gh 50 45 40 35 30 25 20 15 10 5 0 no age 10-19 20-29 30-39 40-49 50+ Alive 22 12 8 5 0 0 Dead 44 2 24 17 5 6 Figure 4.2 Cross Tab of Age Range and Extent of Damage Source: Field Data, 2017 4.1.3 Forms of Attack Different forms of attack were reported for the cases of vigilantism. These attacks include burning, vandalism, lynching/murder, shootings, physical assault and sexual assault. Some victims experienced a combination of these attacks. It was noted that the most dominant form of attack was physical assault, representing 43% of all cases. Victims who suffered this attack were brutalized, beaten or nearly lynched. The next highest attack was actual lynching, representing 37% of cases. Some victims were also burnt or set ablaze, forming 8% of cases. There were instances where the type of attack was not stated. This accounted for 6% of the cases. The remaining attacks were vandalism (2%), shooting (1%) and sexual assault (3%). Most often than not, female victims suffered sexual assault. In addition to being beaten, they were stripped naked 56 University of Ghana http://ugspace.ug.edu.gh with people touching and inserting objects into their private parts. Similarly, a cross-tabulation of the type of attack and survival of that attack was undertaken. Of the 53 victims who were lynched, 51 died at the spot. Also, of the 62 victims who experienced physical assault, 32 survived and 30 died. Victims of sexual assault and vandalism survived the attacks all cases. Victims who suffered burns or shootings all did not survive their attacks Table 4.3 Forms of attack Forms of Attack Frequency Percentage (%) Vandalism 3 2% Burnt 12 8% Lynched/ Murdered 53 37% Physical Assault 62 43% Sexual Assault 5 3% Shot 2 1% Not stated 8 6% Source: Field Data, 2017 4.1.4: Reasons for vigilantism The following themes represent reasons and motivation for perpetrating the act of vigilantism. Most of the reasons given for vigilantism were obtained by journalist from eyewitness accounts and published in the media sites reviewed. Four broad themes were deduced from the online review: (1) Criminal issues; (2) Social issues; (3) Political issues and (4) The justice system. Under criminal issues the researcher found theft/robbery and homicide (murder & accident) to be the main cause of vigilantism. Under social issues witchcraft and sexual orientation were seen as factors leading to the act of vigilantism. Political issues pertain to political turbulence that gave 57 University of Ghana http://ugspace.ug.edu.gh rise to vigilantism acts. These factors lead to people being assaulted if they are perceived to have engaged in or be connected to any of the activities listed under each of the themes i.e. criminal, social and political issues. Criminal issues Most of the reasons cited for the reported acts of vigilantism bother on allegations different forms of crime perpetrated by the victim, which included thefts/robbery and homicide. Most of the vigilantism acts were meted out in response to theft or robberies within communities. Specifically, theft and robbery were cited as the underlying reasons for 74% of the reported cases of vigilantism. In most cases, the victims were reported to have been involved in attempted theft or robbery or caught in these acts. In some cases, victims were reported to have been attacked following suspicions that they were thefts or armed robbers. The following extracts illustrate attributions made to theft and robbery as reasons for engaging in vigilantism: “a pub owner accused someone of stealing her money, raises alarm and residents, mostly men who rushed to the shop will not even ask the victims side of the story but will pounced on the victim and beat him to pulp” (case no.15, adomonline.com, 2017). “Joseph Salifu tried to snatch a bag containing laptop from a student at around 4:30 am at gun point. He was caught by residents, beating and set ablaze” (case no.2, graphiconline.com, 2015). “Likewise, the people of Jirapa went rampage striking down all criminals and ex- convicts when the arrest and judgement of some ruminant and motor bike thefts were not settled to their satisfaction. This lead to the lynching of 4 individuals in the town who were thieves and ex-convicts” (case no.6, graphiconline.com, 2015). “Amina was caught allegedly stealing a laptop and other gadgets belonging to a female student at university of Ghana campus” (case no.62, ghanaweb.com, 2011). 58 University of Ghana http://ugspace.ug.edu.gh Homicide was found to be a notable reason for the occurrence of vigilante acts. In five cases, presenting 3.4%, acts of vigilantism were attributed to murder or suspected murder. In all cases attributed to murder, the victims were suspected or accused of being involved in the murder or attempted murder of another person. The following account illustrate this point: “The Administrator of the Goaso Government Hospital, Mr. Anthony Yeboah Boateng, was on Easter Sunday murdered in cold blood by a group of young men on suspicion that he was part of a criminal gang involved in alleged serial murders in the area. The deceased's attempt to explain to his attackers that he was an innocent person rather inflamed their passions, particularly so when they spotted a corpse in his vehicle. Mr. Boateng, 38, until his death was said to be conveying the corpse of his mother-in-law to the Goaso Government Hospital when the youth clubbed him to death and attempted to set his body and car ablaze. A policeman who tried to save Boateng's life was also clubbed and rushed to a nearby clinic. But that was after he had managed to rescue three other persons in the car — a Catholic nun, the wife of the deceased and a nurse — and rushed them to a local police station” (case no.48, modernghana.com, 2007). “Orga a trader as known and called, was lynched by co-habitors for allegedly stabbing to death a fellow Nigerian over a stolen tooth brush on the Nungua beach Drive road” (case no.81, pulse.com.gh, 2016). In another case, which appears to the outcome of a civil dispute between two sides, the victim of the vigilante act was killed for physically assaulting another person from the opposite side: “A man named Razak was shot dead over a land dispute Coco Beach, Nungua in Accra. He was alleged to be among three armed men who attacked one Bismark Attoh over the piece” (case no.80, pulse.com.gh, 2016). Similarly, in one of the reported cases of vigilantism the victim was alleged to have accidentally knocked down another road user (motor cyclist), and thereby attracted the anger of some youth 59 University of Ghana http://ugspace.ug.edu.gh in the community, who physically assaulted and killed him in the process. The extract below depicts this case: “Residents of Nima, a suburb of Kumasi assaulted and killed a taxi driver by name Awudu for accidentally knocking down a motor rider, who was treated and discharged same day at the Manhyia Hospital” (case no.39, myjoyonline.com, 2017). Social issues Social issues such as witchcraft and sexual orientation (homosexuality or lesbianism) were also some reasons cited for the purported acts of vigilantism in Ghana. Witchcraft cases were 7 in total out of the 145 cases used for the analysis, which represent 4.8%. Most often than not, persons accused of witchcraft were older people, with few being young. Some extracts indicating issues of witchcraft; “A 67-year-old woman accused to be a witch has been stoned to death by the people of Tindongo in the Upper East region. Despite the woman denying being a witch, the community went ahead to kill her by stoning. The people of the community accused the old lady of visiting them in their dreams at night and said she was the cause of several misfortunes in their life. Before the Chief could sit on the matter a mob whisked the old lady and meted instant justice to her by stoning her to death” (case no.87, citifmonline.com, 2017). “A 36 years old famer of Duusi in the Talensi District was accused of witchcraft by enchant children in the area to produce money for him. Neighbors of the village gave him severe beatings until he was rescued” (case no.13, citifmonline.com, 2017). “An elderly woman aged 72, also met her untimely death at Tema site 15 when she was accused of witchcraft by some neighbors of her vicinity. She was beating and set ablaze” (case no.12, ghanaweb.com, 2010). 60 University of Ghana http://ugspace.ug.edu.gh Similarly, in the reported cases, some people suffered vigilantism when they were caught engaging in homosexual or lesbian sex. Three (3) cases out of the 145 cases reported were issues regarding sexual orientation. This forms 2.1% of the entire cases retrieved. Below are some extracts; “Irate youth brutally assaulted a man suspected to be a homosexual at Nima, a suburb of Accra. The victim, identified as Sanusi, was captured in a video footage showing some Nima youth assaulting him while accusing him of being a homosexual in the sprawling Accra suburb” (case no.45, peacefmonline.com, 2015). “A young man in his early thirties was nearly lynched in Tamale, the Northern Regional capital when he was caught in the act with his gay partner. The wife upon seeing the two in the act, raised an alarm for neighbors to come clutching cudgels only to realized that one of the men in the act was a respectable member of the society and other a total stranger who was believed to be a Ghanaian international student in Canada. They were physically assaulted but managed to escape” (case no.72, The Herald, 2014). Political issues The results of the study indicated that, electoral violence and political power results in vigilantism acts. The study found eight (8) cases of political issues from the media sources reviewed. This represented 5.5% of the entire cases reported and retrieved. Political turbulence played out in instances such as; “The Invisible force in June 2017, has assaulted both drivers and passengers of Ashaiman Taboo line Biakoye Trotro station on reason that their party is in power and have been asked to take over management of the lorry station” (case no.31, adomonline.com, 2017). 61 University of Ghana http://ugspace.ug.edu.gh “The delta force also in another incidence went vandalizing the Regional Coordinating Council in Nhyeiso, because they cannot work with George Agyei and that he was 'not part of the struggle' to wrestle power from the NDC during the December 2016 general elections” (case no.33, myjoyonline.com, 2017). This brings us to the end of result presentation for study I. Further discussion of the study is done in the next chapter and recommendations given to how issues of vigilantism should be handled in order to minimize it occurrences or curb it in the country. The next section gives detailed presentation of data and analysis done for the Study II. Justice delivery system The findings revealed that some vigilantism acts within the criminal issues, were as a result of liniment justice system. This is seen in the articles as; “Before embarking on the exercise, the suspects were said to have served notice on the police about their intentions. ASP Afagbedzi explained that the killing of the ex- convicts stemmed from the fact that the people were disillusioned with the justice delivery system in the area. According to him, one of the ex-convicts, Tang, had recently been sentenced to nine months in prison for stealing by the Jirapa Magistrate’s Court. Tang appealed against the sentence and it was commuted to a fine which he duly paid. Some few days later, he was spotted moving about freely in the town, a development which the people did not understand. ASP Afagbedzi said the decision did not go down well with the people, who decided to visit instant justice on all suspected criminals and ex-convicts in the town.” (case no.8, graphiconline.com, 2015). 62 University of Ghana http://ugspace.ug.edu.gh 4.2 Results for Study II 4.2.1 Preliminary analyses This section focuses on the preliminary test done on data set to test for assumptions and make data clean for analysis. Simple frequencies were run on data set, correlations and missing data identified. Factor analysis was also conducted to examine the factor structure of measurement instruments. Thereafter, some variables that were not loading or cross loading were removed before computations were done on the entire data set. These statistics then permitted the researcher to further test for the various hypothesis set in the work. Results are then presented based on socio demographic data and the study’s hypotheses. Data Screening After data collection, data was entered using census survey processing system (CSPro). This tool was used because it makes data entry faster and as well have features that can put constraints on entry variables, example, a variable should not accept values. This feature could not be possible using SPSS. After the researcher enters the data, same questionnaire was given to an independent person to also make data entering. This was done to check for data errors and inconsistencies. Thereafter, the two set of data were compared and the errors and inconsistencies that were flagged corrected to reflect what is on the questionnaire as the respondent put it across. Data set then was exported to SPSS for further analysis. Following the initial screening of the data for errors, missing data analyses were conducted to ascertain the pattern of missing data. The results indicated that the pattern of missing data was characterized as missing at random (MAR) because it can be predictable from other variables in the data set. The output from the analysis gave a few (5% and less) missing in 63 University of Ghana http://ugspace.ug.edu.gh random pattern from the 400 data set we analyzed. According to Tabachnick and Fidell (2007, p. 62), missing values 5% and less will not have serious problems on data especially when they are missing at random and not because of the dependent variable. With this, the researcher assumes it would have no effect on the data set. Exploratory Factor Analysis In addition to computing the alpha coefficient of reliability, the researcher also investigated the dimensionality of the scale by conducting factor analysis. Looking at the total variance explained in the eigen value for all factors extracted is quite good (above 40%) which is good for further analysis. Additionally, all the variables on the scaled used in data collection has a superb Kaiser Meyer Olkin (KMO), the least being .82 and highest being .95. The strength of the relationship among variables indicated by the Bartlett’s Test of Sphericity is significant (0.00) for all variables. Bartlett Test of Sphericity shows validity and suitability of data to the phenomenon under investigation through the study. For a suitable factor analysis to be recommended, the Bartlett’s Test of Sphericity must be less than 0.05 and the present study had 0.00. In fact, the significance level is good enough to reject the null hypothesis. This means that correlation matrix is not an identity matrix. Principal components analysis was used in the extraction method to get the better item loading (Costello & Osborne,2005; Tabachnick & Fidell, 2007). Communality values of the various variables were also good as total variance explained shows how much of the variance in the variables has been accounted for by the extracted factors in all variables. Using the Kaiser criterion, all factors with eigenvalues greater than 1.0 were retained. Only items with factor loadings greater than .30 were retained for each factor (Tabachnick & Fidell, 2007). A few items 64 University of Ghana http://ugspace.ug.edu.gh that poorly loaded on their respective factor or cross-loaded on other factors were deleted. All negatively worded items were reversed before analysis. All variables had one factor extracted except personality scale that extracted two factors. Factor loadings for most of the items were greater than 0.3. As shown in Table 4.4, the amount of variance explained by the extracted factors ranged from 25% to 60%. In general, the principal component analysis was done to explain the covariances or correlations between the variables whiles reducing data into smaller components and as well explain as much total variance in the variables as possible. 65 University of Ghana http://ugspace.ug.edu.gh Table 4.4a Factor loadings and eigen values for the various variables (scales) being used Factor 1 Variance explained Attitude towards vigilantism 1. It is alright for members of the public to beat up crime suspects. .743 2. People who kill armed robbers should not be blamed. .652 3. It is sometimes OK for people to take the law into their own hands if they f .673 47% 4. It is pointless to hand over a suspected criminal to the police because they .759 5. I feel satisfied if criminals are punished by community members than the law .784 6. Beating of thieves is good and should be encourage. .813 7. My mood rapidly cycled in terms of anger and anxiety when criminals are caught. .606 8. I don’t belief in the right of a criminal. .700 9. I believe in punishing those who wrong me or others. .543 10. Punishing criminals out of the legal domain is sensible. .606 13. Beating of criminals deters criminal activities. -.608 14. There should be no restriction on beating criminals caught. .712 66 University of Ghana http://ugspace.ug.edu.gh Table 4.4b Factor loadings and eigen values for the various variables (scales) being used Factor 1 Variance explained Negative reciprocity belief 1. I believe in the saying, “A tooth for a tooth and an eye for an eye”. 2. If someone dislikes me, I will dislike the person too. .632 3. If a person despises me, I will despise them too. .793 55.6% 4. If someone says something nasty to me, I will say something nasty back. .852 5. If a person wants to be my enemy, I will treat them like an enemy. .839 6. If someone treats me badly, I feel I should treat them even worse. .821 7. If someone treats me badly, I will treat that person badly in return. .799 8. If someone important to me does something negative to me, I will do something even more negative to .852 them -.787 9. A person who has contempt for me deserves my contempt. .697 10. When someone hurts me, I will find a way they won’t know about to get even. .505 11. I will not give help to those who treat me badly. .683 12. If you distrust me, I will distrust you too. .591 67 University of Ghana http://ugspace.ug.edu.gh Table 4.4c Factor loadings and eigen values for the various variables (scales) being used Factor 1 Variance explained Police trustworthiness 1. The Ghana police are trustworthy. .803 2. I have confidence in the Ghana police. .859 3. The Ghana police are usually honest. .752 60% 4. The Ghana police always act within the law. .758 5. The police can be trusted to make decisions that are right for the people in my neighborhood. .792 6. People’s basic rights are well protected by the police in my neighborhood. .744 7. The police in my neighborhood are generally honest. .767 8. Ghana has one of the best police forces in Africa. .753 9. I am proud of the work of the Ghana police force. .838 10. I am happy to defend the work of the Ghana police force when talking to my friends. .855 11. I agree with many of the values that define what the Ghana police force. .717 12. I cannot think of another police force that I respect more than the Ghana police force. .691 13. The work of the Ghana police force encourages me to feel good about our country. -.751 68 University of Ghana http://ugspace.ug.edu.gh Table 4.4d Factor loadings and eigen values for the various variables (scales) being used Factor 1 Variance explained Sense of safety 1. Overall, my neighborhood is a good place to live in terms of security. .789 .766 2. There are not many instances of crime in my neighborhood. .715 3. Overall, the police are doing a good job in my neighborhood. .762 53% -.537 4. Crime levels in my neighborhood have changed for the better in the last year .774 5. I feel unsafe being alone. 6. To what extent do you feel safe in your neighborhood? 69 University of Ghana http://ugspace.ug.edu.gh Table 4.4e Factor loadings and eigen values for the various variables (scales) being used Factor 1 2 Variance explained Personality types 1. I am a talkative. .437 2. I usually take the initiative in making new friends. .760 3. I can easily get some life into a rather dull place. .719 25% 4. I like mixing with people. .748 5. I like plenty of movement and excitement around me. .704 6. I am mostly quiet when I am with other people -.738 7. I can get a party or an activity going. .570 8. I sometimes feel ‘just miserable’ for no reason. .650 9. I feel easily hurt. .678 44.5% 10. I am a worrier. .755 11. I often feel lonely. .697 12. I am short-tempered. .635 70 University of Ghana http://ugspace.ug.edu.gh Table 4.4f Factor loadings and eigen values for the various variables (scales) being used Factor 1 Variance explained Religiosity 1. I feel God’s presence. .717 2. I experience a connection to all aspects of life. .666 3. During worship, or at other times when connecting with God, I feel joy which .769 4. I find strength in my religion or spirituality. .794 52% 5. I find comfort in my religion or spirituality. .813 6. I feel deep inner peace or harmony. .747 7. I ask for God’s help in the midst of daily activities .790 8. I feel guided by God in the midst of daily activities. .808 9. I feel God’s love for me, directly. .793 10. I feel God’s love for me, through others. .702 11. I am spiritually touched by the beauty of creation. .780 12. I feel thankful for my blessings. .777 13. I feel a selfless caring for others. .607 14. I accept others even when they do things I think are wrong. .615 15. I desire to be closer to God or in union with Him. .765 16. In general, how close do you feel to God? .349 71 University of Ghana http://ugspace.ug.edu.gh 4.2.2 Descriptive statistics and correlations Tests of assumptions of multivariate analysis This was done to check normality of distribution and homogeneity of variance. The data set indicated a normal distribution and no outliers are present. Meaning the points are randomly distributed about the value of 0. The variables gave a well descriptive statistics with a good skewness and kurtosis scores. Data was normally distributed and Bartlett’s test for sphericity showed a favorable statistically significant results, indicating all instrument used to be factorable. Actually, except religiosity, which was slightly skewed, all the measures were normally distributed. Religiosity scale scored a kurtosis of 3.17 and an observation of univariate outlier was identified, removed and analysis run again to get a good skewness and kurtosis as shown below. Table 4.5 Descriptive statistics of the various variables Variables Mean Std. Skewness Kurtosis Reliability Dev. . Negative reciprocity 2.49 .68 57 .41 .86 Attitude to Vigilantism 2.61 .71 .49 .03 .83 Police trustworthiness 2.64 .67 -.09 -.52 .89 Sense of safety 3.08 .62 -.04 .03 .63 Extraversion personality 3.19 .52 -.29 .18 .67 Neuroticism personality 2.81 .63 .23 -.01 .75 Religiosity .08 .84 -.78 .47 .94 Source: Field Data, 2017 72 University of Ghana http://ugspace.ug.edu.gh Reliability and Correlation Reliability test was conducted using Cronbach alpha to test whether the scales used were reliable and will measure exactly what it intends to measure. The alpha coefficient for the items on the attitude scale is .83, negative reciprocity belief scale (α=.86), sense of safety scales (α=.63), police trustworthiness scale (α=.89), religiosity (α=.94) and personality scale (extraversion α=.67 and neuroticism α=.75), suggesting that the items have relatively high internal consistency. A symmetrical correlation was given by the data set. The item–constructs, indicated an acceptable level of consistency (Tabachnick &Fidell, 2007). The data set gave a linear relationship between variables as indicated in Table 4.6. The items correlated very well with each other except experienced of crime which did not have any relationship with the other variables. 73 University of Ghana http://ugspace.ug.edu.gh Table 4.6 Correlation table 1 2 3 4 5 6 7 8 9 10 11 12 1. Age 1.00 2. Sex -.06 1.00 3. Marital status -.62** .02 1.00 4. Religion .06 -.02 .01 1.00 5. Employment status .51** .00 -.40** .05 1.00 6.Negative reciprocity -.05 -.06 -.09* .11* -.02 1.00 7. Attitude support .09* -.09 -.17** .10* .11* .58** 1.00 8. Police trust .23 -.14 .03 -.07 -.19** -.18** .4** 1.00 9. Sense of safety .03 -.08 .14** -.02 -.11* -.23** .16** .14** 1.00 10.Extraversion -.02 -.07 .04 -.15** .00 -.04 -.20 .14 .12 1.00 11. Neuroticism -.14** .09* .03 .03 -.11* .25** .25** -.05 -.05 -.09* 1.00 12. Religiosity .16** .12 -.05 -.01 .13** -.13** -.19** .17** .12** .29** -.21** 1.00 13. Experience of crime .06 -.02 -.07 .04 .06 -.01 .04 .00 -.02 -.08 -.04 -.01 ** Correlation is significant at the 0.01 level (1-tailed). * Correlation is significant at the 0.05 level (1-tailed). (**) p < .01 & (*) p < .05 Source: Field Data, 2017 74 University of Ghana http://ugspace.ug.edu.gh 4.2.3 Hypothesis testing Hierarchical multiple regression analysis was conducted after all assumptions were met from the data set. This type of regression allows for specific fixed order of entry for variables in order to control for the effects of covariates or to test the effects of predictors independent of the influence of others. The outcome variable for the test is attitude towards vigilantism, conceptualized as support for vigilantism. The predictor variables were group into three: demographic variables (age, sex, marital status, and employment status), situational factors (police trustworthiness, sense of safety, and experience of crime) and personal factors (religiosity, negative reciprocity beliefs, and personality dimensions). In running the analysis, the outcome variable was first regressed on the demographic variable, followed by situational factors, and then personal factors. A summary of the results for the hierarchical multiple regression analysis is presented in Table 4.7. As shown in Table 4.7, overall, the model predicting support for vigilantism was significant, with predictors accounting for 43% of the variance in attitudes toward vigilantism. At step 1 of the model the demographic variables significantly accounted for 6% of the variance in attitude toward vigilantism. When entered at step 2, the situational variables accounted for an additional 6% beyond the demographic variables. Finally, at step 3 of the model personal factors explained additional 31% of the variance on attitudes towards vigilantism beyond the demographic variables and situational factors. The study’s hypotheses were tested based on the standardized regression coefficients for the variables are step 3 of the models. 75 University of Ghana http://ugspace.ug.edu.gh Table 4.7 Regression model for demographics on attitudes towards vigilantism, situational factors on attitudes towards vigilantism, and personal factors on attitudes towards vigilantism Model I Model II Model III Variables B SE β B SE β B SE β Constant 3.09 .29 3.71 .33 1.92 .41 Demographics Age -.04 .05 -.05 -.02 .05 -.03 .05 .04 .07 Sex (male vs female) -.10 .08 -.07 -.10 .08 -.07 -.08 .06 -.06 Marital status (single vs married) -.27 .10 -.18** -.24 .10 -.15 -.07 .08 -.05 Religion (Christians vs Muslims) .12 .06 .10 .11 .06 .09 .03 .05 .02 Employment status (Employed vs Unemployed) .14 .09 .09 .10 .09 .07 .15 .07 .10** Situational factors Police trustworthiness -.16 .06 -.15** .03 .05 -.03 Sense of safety -.11 .06 -.10** -.04 .05 -.04 Experience of crime _ victim .27 .51 .07 .35 .42 .09 Experience of crime _ witness .03 .12 .03 -.06 .10 -.07 Personal factors Extraversion -.20 .06 -.15** Neuroticism .13 .05 .11** Religiosity -.04 .04 -.04 Negative reciprocity belief .53 .05 .51** Model summary R2 .06 .12 .43 Change in R2 .06 .06 .31 F for change in R2 4.18* 6.02** 46.84** Adjusted R2 .04 .09 .41 F 4.18* 5.13** 19.82** Df 5,355 9,351 13,347 p< .05**. p< .01*. A- Employed, B-Unemployed. Source: Field Data, 2017 76 University of Ghana http://ugspace.ug.edu.gh The first set of hypotheses (Hypothesis 1 and Hypothesis 2) suggested that demographic variables (age and sex) would have significant relationships with attitudes towards vigilantism. The results of the hierarchical multiple regression analysis showed that age (β = .07, p > .05) had no significant relationship with attitudes towards vigilantism. This shows that Hypothesis 1, which stated that age would have a significant relationship with attitudes toward vigilantism, was supported. Similarly, sex (β = –.06, p > .05) was not a significant predictor of attitudes towards vigilantism. Thus, Hypothesis 2, which stated that sex would be significantly related to attitudes towards vigilantism, was not support. In other words, there was no significant difference between males and females in terms of their attitudes towards vigilantism. Although not hypothesized, employment status (β = .10, p < .01) was significantly related to attitudes toward vigilantism. Thus, employed persons were more likely to support vigilantism than unemployed persons. Religion (β = .02, p > .05) and marital status (β = –.05, p > .05) were not significantly related to attitudes toward vigilantism. The second set of hypotheses (Hypotheses 3, 4, and 5) suggested that situational factors would have significant relationships with attitudes towards vigilantism. However, although the situational factors accounted for a significant amount of variance in attitudes towards vigilantism at step 2 of the model, they fail to account for significant amount of variance at step 3. As shown in Table 4.7, police trustworthiness (β = –.03, p > .05) was not a significant predictor of attitudes towards vigilantism. Thus, Hypothesis 3, which stated that police trustworthiness would be significantly related to attitudes toward vigilantism, was not supported. Similarly, sense of safety (β = –.04, p> .05) did not significantly predict attitudes toward vigilantism. Therefore, Hypothesis 4, which stated that sense of safety would be negatively related to attitudes towards vigilantism, was not supported. Likewise, being a victim of crime (β = .09, p > .05) or 77 University of Ghana http://ugspace.ug.edu.gh witnessing a crime (β = –.07, p > .05) did not significantly predict attitudes toward vigilantism. Therefore, the hypothesis that experience of crime would be positively related to attitudes toward vigilantism (Hypothesis 5) was not supported. The final set of hypotheses (Hypotheses 6, 7, and 8) suggested that personal factors would be significantly related to attitudes toward vigilantism. The results, as shown in Table 4.7, indicated that the personality dimension of extraversion (β = –.15, p < .01) was a significant negative predictor of attitudes toward vigilantism. Conversely neuroticism (β = 11, p < .05) was a significant positive predictor of attitudes toward vigilantism. Thus, individuals who scored higher on extraversion reported lower levels of support for vigilantism, while those who scored higher on neuroticism reported higher levels of support for vigilantism. These results provide partial support for Hypothesis 6, which stated that extraversion and neuroticism would be positively related to attitudes toward vigilantism. Contrary to expectation, religiosity (β = –.04, p > .05) was not significantly related to attitudes toward vigilantism. Thus, Hypothesis 7, which stated that religiosity would have a negative relationship with attitudes toward vigilantism, was not supported. In other words, levels of religiosity were not significantly associated with support for vigilantism. Finally, negative reciprocity belief (β = .51, p < .01) had a significant positive relationship with attitudes towards vigilantism. Thus, high levels of negative reciprocity beliefs were associated with greater support for vigilantism. This result provides support for Hypothesis 8, which stated that negative reciprocity belief would be positively related to attitudes towards vigilantism 4.2.4 Chapter summary The chapter presented results on both the archival study (Study I) and the survey (Study II). The overall results suggest more vigilantism cases being reported in the main stream media, with 78 University of Ghana http://ugspace.ug.edu.gh people’s attitudes towards vigilantism being predicted more by negative reciprocity belief and the neuroticism personality types. The next chapter (5) presents a discussion of the findings together with their implications in the light of previous research and literature reviewed. Recommendations are given and the entire work is concluded with a summary and conclusion. 79 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE DISCUSSION, RECOMMENDATIONS AND CONCLUSION 5.0 Overview This study investigated vigilantism in Ghana by examining the patterns, nature and scope of vigilantism, as well as attitudes towards vigilantism. The study undertook a review of media reportage on vigilantism, and a survey of public attitudes towards vigilantism. This present chapter concludes the study by giving a discussion of the study, its limitations, recommendations and implications for research and policy. A conclusion is then given to bring the entire study to a close. 5.1 Discussion of research findings for Study I Study I undertook a situational analysis of vigilante violence. This study analyzed the content of media reports on vigilantism in Ghana by answering the research question: What is the scope, nature and patterns of vigilantism in Ghana? The analysis showed six main findings: (1) that incidents of vigilantism are on the increase across the regions of the country; (2) the rise in vigilantism is mainly from the urban areas of the country with Greater Accra being the highest; (3) majority of the victims that experience any form of vigilantism died; (4) youngsters (10 to 19 years) and the aged (50 years above) are not spared either from the menace of vigilantism in Ghana; (5) most of the victims of the recent study were in gainful employment either as self- employed persons or employees; and (6) the reasons for vigilantism were related to criminal issues, social issues such as witchcraft and homosexuality, and political issues. 80 University of Ghana http://ugspace.ug.edu.gh 5.1.1: Patterns of vigilantism A major finding in the study relates to the trend of increasing vigilantism in Ghana. As opposed to 48 cases of vigilantism which were reported in Adinkrah’s (2005) study, a total of 145 cases were recorded in this present study. The increase in trend could be due to the fact that more vigilantism cases are being reported in the mainstream media. It could also be that more people are resulting to vigilantism as a self-help to gaining justice. Thus, as some of the articles explained that the justice system is liniment on criminals and their punishment does not equate the crime or offenses committed, they will resort to vigilantism as a way of getting justice they deserved. The results from this present study suggest that males form the largest group (89%) of people who fall victim to vigilantism. Compared to Adinkrah’s work of 2005 where females constituted 11% of vigilante victims, the current study recorded 8% for female cases. This could be due to the fact that society may either be lenient to female offenders, or that females are less likely to have engaged in criminal activities. The assertion of lesser female involvement in crime is evidenced in the literature. As has been indicated by Newburn and Stanko, (1994), Heimer and De Coster (2001), and Steffensmeier and Allan (1996), greater number of males commit crimes compared to females. As vigilante violence is perpetrated primarily in response to crime, males become more likely to suffer such violence compared to females (De Coster, Heimer & Wittrock, 2006). This could be a key explanation for the greater incidence of vigilante violence occurring to males as opposed to females. In this present study, vigilante violence was found to occur predominantly to persons aged 20 to 30 years, though other age groups suffered a similar fate. As elucidated in earlier studies by Outwater, Mgaya and Campbell, (2013), persons aged 20 to 29 years tended to suffer 81 University of Ghana http://ugspace.ug.edu.gh most in terms of death that mainly occurred in the open (Brodie, 2013; Goldbaum, 2014). Similarly, uniform reports from FBI arrest data (1935 – 1997) indicate a long-term trend of younger age-crime and the peak age-crime involvement being younger than 25 for all reported crime. The trend of more youthful people suffering vigilantism can be linked to the incidence of higher youthful involvement in crime; a phenomenon which has come to be known as the “age- crime curve” (Rocque, Posick & Hoyle, 2015). The “age-crime curve” indicates that high rates of crime are perpetrated during the age of 14 years, peaks in the early to mid-20s, and then declines thereafter. This may possibly explain why majority of victims in this study were found to be aged 20 to 30 (Boateng, Olumide, Isabalija & Budu, 2011). 5.1.2: Scope of vigilantism Greater Accra recorded the highest cases of vigilantism followed by Ashanti region. These acts of vigilantism are seen more in the urban areas of the various regions. Vigilantism is thought to occur throughout the ten (10) regions in Ghana and mostly in the streets, public places, and residences, with some occurring in market places (Outwater, Mgaya, & Campbell, 2013). In this present study, 75% of the incidence of vigilantism occurred in urban centers whiles the rural areas recorded 25% of cases. This is confirmed in the literature, where vigilantism is seen to dominate in urban areas; though rural areas also see some incidence of the phenomenon occurring. In Kenya, for example, the Mungiki group is a known vigilante group that operates in the rural area (Anderson, 2002; Kagwanja, 2003; 2005; Wamue, 2001). Studies have shown higher levels of crime in urban areas compared to the rural areas (Bruinsma, 2007; Salzani, 2007; Sampson & Groves 1989). According to Salzani, (2007) and Krivo and Peterson (1996), higher urban crime rates are as a result of lower levels of social cohesion and informal social control. 82 University of Ghana http://ugspace.ug.edu.gh The higher crime rates in urban areas would then mean that, greater chances exist for vigilantism to occur in these areas, as vigilantism is both a response and form of self-help against crime. Vigilantism was also noted to have occurred in various forms, such as burning, vandalism, lynching/murder, shootings, physical assault and sexual assault. In the literature as well, vigilantism has been reported to take forms such as physical assault, “necklacing” (where victims are set ablaze using car tyres) (Harnischfeger, 2003; Minnaar, 2001), and burning and stoning (Ng'walali & Kitinya, 2004). 5.1.3: Motivations for vigilantism The findings of this study show crime to be a major reason for the meting out of vigilante violence. This finding is corroborated in the literature, where vigilantism has been indicated to be dispensed in response to crime and for community protection (Harnischfeger, 2003; Minnaar, 2001; Ng'walali & Kitinya, 2004; Outwater, Mgaya, & Campbell, 2013). Furthermore, witchcraft was found to be a reason for the occurrence of vigilantism. It is generally believed that witches have supernatural powers they employ in causing harm; knowingly or unknowingly, to other people, or for their own personal benefit (Nukunya, 2004). According to Schnoebelen (2009), the accusation of witchcraft is a weapon of control against women in some cultures. Evans Pritchard (1976) has indicated that witches can harm others by virtue of an inherent ability they possess. Nukunya (2004) and Pritchard (1937) aver that witchcraft is generally believed to harm people and cause destruction in a person. Furthermore, witchcraft is believed to be associated with evil spirits that only demand bad things for people they come in contact with, as such are danger to society. Similarly, people with anti-social habits, excessive wealth or success, abject poverty and wretchedness may be branded as witches. 83 University of Ghana http://ugspace.ug.edu.gh The reasoning behind this assertion is that excessive wealth could only come about through witchcraft, while acute poverty could be used as a camouflage in hiding the identify of a witch (Nukunya, 2004). Sexual orientation was also identified as a motivating factor for vigilantism in this study. One of the attributions people gave for vigilante violence was that the victim was believed to be a homosexual or a lesbian. The Ghanaian culture frowns on homosexuality or any kind of attraction between persons of the same sex. The public is ready to assault anyone caught engaging in any form of homosexuality. Though a crime in Ghana, homosexuality is practiced in various parts of the country. Generally, most Ghanaians consider themselves as heterosexual (Appiah, 2006). Perceptions towards homosexuality have been one of total ambivalence in times past, to agitation and condemnation in more recent times. Several studies indicate that the greatest opponents of male homosexuality are people with high religious inclinations and who attend service regularly. They generally adhere to conservative religious dogma and ideology (Alston, 1974; Bhugra, 2010; Cameron & Ross, 1981; Herek, 1984; Nyberg & Alston, 1977). Similarly, negative reactions towards male homosexuality have been argued to imply a defensive mechanism for reinforcing masculine stereotypes (Black & Stevenson, 1984). People with anti- homosexual tendencies consider themselves to be better off and think of homosexuals as sick people (Bleek, 1976; Kan Au, Chan, Cheung, Lam, Liu & Wong, 2009; Lumby, 1976; Osei, 2011). People who possess positive attitudes towards homosexuals are more likely to know and relate with homosexuals personally (Bowman, 1979; Glassner & Owen, 1976; Milham, San Miguel & Kellogg, 1976). In this present study, some people were found to suffer vigilantism when they were caught engaging in homosexual or lesbian sex. 84 University of Ghana http://ugspace.ug.edu.gh In this study, political turbulence was noted to give rise to vigilantism in Ghana. Ghana has experienced several years of military rule after independence. With the excption of Nkrumah who experienced six years of uninterrupted republican rule (1960-1966), most other republics only lasted for up to three years, after which they were overthrown in a coup d’état. The close to a decade of quasi–military rule by the Provisional National Defense Council (PNDC) gave way to the adoption of a republican constitution in 1992, and a return to multiparty party democracy in 1993. Ghana has since held seven successful multiparty elections. Indeed, Ghana’s democratic transition and journey can be considered as one of Africa’s political success stories (Gyimah-Boadi, 2008; Ninsin, 1998; Whitfield & Jones, 2007). That notwithstanding, the search for political power by the nation’s political parties has led to the existence of some violence in parts of the country. The way in which power is exerted upon a group can have repercussive outcomes for popularity. This has influenced political leaders to resort to violence in winning elections. Thus, in Ghana, a number of vigilante acts were perpetrated in the search for power by political parties. Some groups which engaged in these acts include the Invincible Forces of the NPP, Azorka Boys of the NDC and the Delta Fore of the NPP. 85 University of Ghana http://ugspace.ug.edu.gh 5.2 Discussion of research findings for Study II The main objective of Study II was to examine public attitudes towards vigilantism in Ghana. The outcomes of the various hypotheses tested in this study found both support and rejection in the literature. The first hypothesis showed no significant association between age and sex on support for vigilantism. First, age was found not to be statistically significant in the present study. This finding is supported in the literature by Haas (2010), who also found age not to be a predictor of vigilantism in the Netherlands. However, Tankebe (2009) found age to be a significant predictor in Ghana. Age not being statistically significant predictor of attitudes toward vigilantism in this study implies that support for vigilantism is not confined to a specific age group, but cuts across all age groups. The present study would not argue that age does not have effect on crime but rather vigilantism is a subset of crime which age cannot relate positively to. In addition to the above hypothesis, which postulated sex as having a significant relationship with attitudes towards vigilantism, was also not supported. Thus, people’s attitudes towards vigilantism were not associated with their sex. Again, the study by Haas (2010), also did not find any support for sex in relation to attitudes towards vigilantism. However, studies by Adinkrah (2005) and Tankebe (2009), observed that men supported vigilantism, since they perpetrated the act compared to women. In addition, Sommera, Likindikokib, and Kaayab (2013) found sex to be significant predictor for attitude towards vigilantism. In the present study, sex was not found to be a significant predictor of support for vigilantism, this can be an indication that both men and women hold similar views regarding matters such as vigilantism in Ghana. Secondly the hypothesis on the situational factors were also not supported. it was hypothesized that police trustworthiness will have a significant relationship with attitudes 86 University of Ghana http://ugspace.ug.edu.gh towards vigilantism. The findings of the study revealed that in the absence of personal factors that give rise to vigilantism, police trustworthiness will be supported and will match the empirical findings of some literature (Jackson, Asif & Bradford 2014; Tankebe, 2009). Again it is believed that persons who perceived the police not to be trustworthy were more likely to support vigilantism (Hough et al., 2010; Jackson et al., 2014; Tankebe, 2013; Tyler 2006). In the presence of other predictor variables, the hypothesis was not supported, as no statistical significant relationship was found between police trustworthiness and attitudes towards vigilantism. Contrary to what other empirical studies are finding and in support to the findings of the present study, Haas et. al (2012) found that support for vigilantism was not due to lack of confidence in the legitimacy of the criminal justice system, but rather the extent or severity of the crime committed. This implies that public support for vigilantism may be ensuing from the severity of crime committed in the society and not the lack of trust in the police. Furthermore, no significant relationship was found between previous experience of crime and support for vigilantism in the present study. Although experienced of crime is estimated to predict violence, evidence in literature connecting this is limited. Some literature may find association between experiencing unjust harm and wanting revenge to be consistent with the view that revenge is basically driven by a desire to restore a personal sense of justice (Frey, Pearson, & Cohen, 2015; Osgood, 2017). According to Pedersen, Kurzban, and McCullough (2013), vergence is selfishly sought for and not because of one’s experience of harm. We can therefore infer that Pederson et. al (2013) study support the present study’s finding. In addition, there was no statistically significant association between sense of safety and support for vigilantism. 87 University of Ghana http://ugspace.ug.edu.gh There are limited studies in showing the relationship between sense of safety and violence or vigilantism (Hartinger-Saunders, Robin, Rine, Nochajski, & Wieczorek, 2012; Migliorini, Rania, Cardinali, & Manetti, 2008). The study by Hartinger-Saunders, Robin, Rine, Nochajski, and Wieczorek (2012) found significant relationship between neighborhood safety and violence. Thus, the hypothesis on sense of safety would have been supported if personal variables were excluded from the study. The last set of hypotheses were on personal factors and indicated statistically significant relationship to attitudes towards vigilantism, except for religiosity which had no statistically significant association with attitudes towards vigilantism. A significant association was found between extraversion and neuroticism personality types and attitudes towards vigilantism. The present study indicated that extraversion personalities are less likely to indulge in vigilantism. In personality and aggression literature, Mccullough, Bellah, Kilpatrick, and Johnson (1998), also found no relation between extraversion personality and aggression. Furthermore, a positive relationship was found between neuroticism personality and attitudes towards vigilantism. These findings are supported in the literature. Bushman and Baumeister (1998) aver that certain personality traits can influence aggressive behavior. Also, neuroticism has been found to be a strong predictor of anger and hostility (Anderson & Huesmann, 2003; Hofmans, Kuppens, & Allik, 2008; Sharpe & Desai, 2001), which may in turn lead to support for vigilantism. Mccullough et. al (1998) have also found neuroticism to be positively correlated to vengeance. In line with expectation, results from the present study provided support for the hypothesis that negative reciprocity beliefs would positively relate with attitudes towards vigilantism. This finding parallel that of Mitchell and Ambrose (2007) who found that negative reciprocity beliefs strengthened the relationship between abusive supervision and supervisor- 88 University of Ghana http://ugspace.ug.edu.gh directed deviance. It appears that negative reciprocity beliefs may promote endorsement of aggressive behaviour, one of which is vigilantism. Gouldner (1960) noted that individuals may endorse a negative norm of reciprocity, under which unfavourable treatment promotes “not the return of benefits but the return of injuries” (p. 172). Cropanzano and Mitchell (2005) also indicate that individuals may be guided by negative reciprocity where they believe that when someone mistreats them, it is acceptable to retaliate in return. Individuals who endorse negative reciprocity believe in retribution as the correct and proper response to unfavorable treatment (Eisenberger, Lynch, & Aselage, 2004), and are thus more likely to seek retaliation than avoidance (McLean Parks, 1998). Findings from the present study failed to support the hypothesis that religiosity would have a negative relationship with attitudes towards vigilantism. Specifically, support for vigilantism was found to be unrelated to levels of religiosity. A possible reason for this finding may be that all respondents are coming from a religious background and most of the religions preach forgiveness of sin and as well as asserts punishment for wrong doers. Messner et al., (2006) and Jacobs et al., (2005) argue that a positive relationship exists between the share of local religious adherents who are fundamentalist Christians and the imposition of the death penalty but fail to find a racial effect. Religion has however been unsuccessfully used in attempting to measure or predict contemporary homicide rates (Jacobs & Schuetze, 2011). Jacobs et al., (2005) and Jacobs and Schuetze (2011) sought to examine whether a county’s religious composition impacted its frequency of lynching, and found that a pluralistic religious marketplace with competing religious denominations weakened the bonds of a cohesive moral community, and might have enhanced white racial solidarity and lower incidence of lynching in counties where a larger share of church members belonged to denominations with racially mixed 89 University of Ghana http://ugspace.ug.edu.gh denominations, suggesting that cross-racial solidarity served to reduce racial violence (Bailey & Snedker, 2011; Jacobs et al.,2005; Jacobs & Schuetze, 2011). Therefore, it is possible that the statistically non-significant results recorded for religiosity in this present study could be due to the fact that Ghanaians do not share the same religious ideologies, as they practice different religions. As some will preach punishment for certain wrong doing, others will preach forgiveness and vergence is for God Almighty. 5.3: Implications for Theory, Research and Practice The present study raised concern for further research on acts of vigilantism in Ghana, by examining the justice delivery system, and convicted perpetrators if any in terms of the motivations for committing such acts as done elsewhere (Haas et. Al, 2012; Tankebe, 2010). Other future primary studies may implore the views of surviving victims of vigilantism to understand the effects of vigilantism. As well, a primary research is needed to investigate why higher number of vigilantism cases are recorded in the rural areas (Gerban, 2007; Krivo & Peterson, 1996). One of the key findings in the present study was that personal factors were, overall, more important in predicting attitudes towards vigilantism, whiles situational and demographic factors were not significant in predicting attitudes towards vigilantism. This key finding is important both for research and practice. A study is needed further to understand why situational and demographic factors were not significant predictors of attitudes towards vigilantism, and also why personal factors predisposes an individual to support for vigilantism. Generally, a holistic approach needs to be adopted in dealing with vigilantism. Studies must be commissioned, which examine various angles to the matter, including the perspectives of perpetrators, law enforcers, the judiciary, surviving victims and their families, and the general public with the goal of 90 University of Ghana http://ugspace.ug.edu.gh ascertaining the basis of vigilantism to allow for policy formulation and implementation. The evidence from these primary studies may inform intervention and prevention efforts in Ghana. Practically, the findings of the present study underscore the significant need to revisit the criminal justice system regarding lenient sentencing and incorporate punishments that will be commensurate with the criminal offence in order to enhance the legitimacy of the justice system (Sekhonyane & Louw,2002; Tankebe, 2010). It was evident from this study that people resort to vigilantism on the basis of criminals obtaining soft punishments or lenient sentencing which does not deter them from committing the same crimes when they are released. Vigilantism, with its blatant disregard for human life and dignity, may serve to weaken the conscience of both perpetrators and onlookers, thereby trivializing human life and making violence more acceptable in Ghanaian society. This can also have negative effects on the criminal justice system by influencing people to resort to violence instead of sending issues to the court for redress. 5.4: Limitations and Recommendations for Future Research Like any other study, the current study faced some challenges and limitation. First the archival study had a limitation regarding missing data. Majority of the cases used for the study lack information mostly on perpetrators and the steps taken by public authorities and security agencies after reports of a mob attack have been lodged with them. Those that indicated an arrest of the perpetrators also failed to capture what was done to these perpetrators after their arrest. Secondly, the study is limited in the demographics proportion of respondents surveyed (Study II). Thus, there were more men than women, and more Christians than Muslims. In as much as possible that the researcher wanted to have varied opinions towards attitudes to 91 University of Ghana http://ugspace.ug.edu.gh vigilantism, it is possible for one ethnic group to dominate than the other. This could have skewed the study towards a direction that the researcher may not be aware of. In addition, the study is limited by its cross-sectional nature which makes it impossible to make causal inferences. Longitudinal studies of public attitudes towards vigilantism are therefore recommended. As well, the media analysis could have been supported by actual police data to get a much accurate reported cases. Given that the two dimensions of personality included in the study both had significant relationships with attitudes towards vigilantism, future studies could examine whether the other three personality dimensions may be related to support for vigilantism Although the study sought to be as exhaustive as possible, there are still other dimensions of the vigilantism dynamics that were not covered in the study. To this extent, the following recommendations are made for further research to allow for a more comprehensive outlook of the issues regarding vigilantism in Ghana. The first recommendation is that, further research should be undertaken on the perpetrators and survivors of vigilantism in order to get a holistic understanding of the phenomenon. It is possible that only selected cases are reported in the mainstream media and others do not. It is therefore recommended that police reports should also be surveyed for further studies. Also, the impact of vigilantism on individuals, family and the nation as a whole should be ascertained in future research 5.5: Conclusion The study examined the scope, nature and patterns of vigilantism as well examined other factors that influence peoples’ attitude towards vigilantism in Ghana. One hundred and forty-five (145) cases were retrieved online and examined using content analysis whiles four hundred respondents were surveyed for the quantitative part of the study. The findings of the study 92 University of Ghana http://ugspace.ug.edu.gh revealed an increased pattern in the trends of vigilantism with causative factors being theft/robbery, witchcraft, sexual orientation, and political turbulence, among other factors. All categories of age groups suffered vigilante acts in the study, with the negative reciprocity belief being found to be a significant predictor of vigilantism. The study represents the first research attempt aimed at presenting some useful insight into vigilantism in Ghana and underscores the need for further primary research efforts on the phenomenon in order to build a nuanced understanding of the key characteristics of the victims and survivors, perpetrators and the act of vigilantism in Ghana. This study has shown that the patterns, scope, nature and attitudes towards vigilantism in Ghana are fairly and comparatively similar to those found in other parts of Africa and the globe. To conclude, it is recommended that further studies be undertaken which properly address the limitations, thereby giving a more holistic and comprehensive view of the issue of vigilante violence in Ghana. 93 University of Ghana http://ugspace.ug.edu.gh REFERENCES Abusuafm, (2017, June 9). “Invisible Forces seize 'trotro' station, assault drivers in Accra” Adom online. Retrieved from http://adomonline.com/ghana-news/invincible-forces-seize-trotro- station-assault-drivers-accra/ Act 29 of Ghana. (1960). The criminal code of Ghana. Accra, Ghana: Ghana Publishing Company Limited, Assembly Press. Adamu, Fatima L. (2008). Gender, Hisba and the Enforcement of Morality in Northern Nigeria. Africa, 78 (1): 136–52 Adinkrah, M. (2005). Vigilante homicides in contemporary Ghana. Journal of Criminal Justice, 33(5), 413-427 Adinkrah, M. (2014). Intimate partner femicide–suicides in Ghana: Victims, offenders, and incident characteristics. Violence against women, 20(9), 1078-1096. Adjei, k. (2017, June 4). “Hot audio: Man, 55 lynched for ‘stealing’ GH¢1, 50 pesewas” Adom online. Retrieved from http://www.adomonline.com/ghana-news/hot-audio-man-55- lynched-stealing-gh%C2%A21-50-pesewas/ Adu-Gyamfi, E. (2014). Effective revenue mobilisation by districts assemblies: A case study of Upper Denkyira East Municipal Assembly of Ghana. Public Policy and Administration, 2(1), 97-122. Alemika, E., & Chukwuma, I. C. (2005). Criminal victimization and fear of crime in Lagos Metropolis, Nigeria. Cleen Foundation Monograph Series, 1. 94 University of Ghana http://ugspace.ug.edu.gh Alston, J. P. (1974). Attitudes toward extramarital and homosexual relations. Journal for the Scientific Study of Religion, 13(4), 479-481. Amnesty international public statement. (2017), 2013(June), 22–23. Ampratwum-Mensah, A. (2007, April 10). “Hospital Administrator Murdered In Cold Blood”. Modern Ghana.com. Retrieved from https://www.modernghana.com/news/133658/hospital- administrator-murdered-in-cold-blood.html Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of Psychology, 53(1), 27–51. Anderson, C.A., & Huesmann, L.R., (2003). Human aggression: a social-cognitive view. In: M.A. Hogg & J. Cooper (Eds.), Handbook of Social Psychology (pp. 296 – 323). London: Sage Publications. Anderson, D. M. (2002). Vigilantes, violence and the politics of public order in Kenya. African affairs, 101(405), 531-555. Appiah, K. A. (2006). The politics of identity. Daedalus, 135(4), 15-22. Atuobi, S. M. (2008). Election-related violence in Africa. conflict trends, 2008(1), 10-15. Awuni, F. ( 2017, June 1).“Family of lynched ‘witch’ blames Pelungu chief”. Citifmonline.com. Retrieved from http://citifmonline.com/2017/06/01/family-of-suspected-lynched-witch- blames-pelungu-chief/ Awuni, F. ( 2017, June 2). “Man accused of being wizard nearly lynched” Citifmonline.com. Retrieved from http://citifmonline.com/2017/06/02/man-accused-of-being-wizard-nearly- lynched/ 95 University of Ghana http://ugspace.ug.edu.gh Bailey, A. K., & Snedker, K. A. (2011). Practicing what they preach? Lynching and religion in the American South, 1890–1929. American Journal of Sociology, 117(3), 844-887. Baloyi, E. (2015). The Christian church's role in the escalating mob justice system in our black townships-An African pastoral view. HTS Theological Studies, 71(2), 01-07. Bayart, J. F., & Ellis, S. (2000). Africa in the world: a history of extraversion. African affairs, 99(395), 217-267. Bazenguissa-Ganga, Rémy. (1999b). The Spread of Political Violence in Congo- Brazzaville. African Affairs, 98 (390), 37–54. Bhugra, D. (2010). Commentary: Religion, religious attitudes and suicide. International journal of epidemiology, 39(6), 1496-1498. Black, K. N., & Stevenson, M. R. (1984). The relationship of self-reported sex-role characteristics and attitudes toward homosexuality. Journal of Homosexuality, 10(1-2), 83- 93. Blalock, H. M. (1967). Toward a theory of minority-group relations. Bleek, W. (1976). Sexual Relationships and Birthcontrol in Ghana a Case Study of a Rural Town. Boateng, R., Olumide, L., Isabalija, R. S., & Budu, J. (2011). Sakawa-cybercrime and criminality in Ghana. Journal of Information Technology Impact, 11(2), 85-100. Bowman, R. (1979). Public attitudes toward homosexuality in New Zealand. International review of modern sociology, 229-238. 96 University of Ghana http://ugspace.ug.edu.gh Brodie, N. (2013). Guide: Understanding crime statistics in South Africa–what you need to know. Africa Check. Bruce, D., & Komane, J. (2000). Taxis, cops and vigilantes: police attitudes towards street justice. Centre for the Study of Violence and Reconciliation. Bruinsma, G. J. (2007). Urbanization and urban crime: Dutch geographical and environmental research. Crime and Justice, 35(1), 453-502. Bushman, B. J., & Baumeister, R. F. (1998). Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence?. Journal of personality and social psychology, 75(1), 219. Buur, L. (2008). Democracy & its discontents: vigilantism, sovereignty & human rights in South Africa. Review of African Political Economy, 35(118), 571-584. Cameron, P., & Ross, K. P. (1981). Social psychological aspects of the Judeo-Christian stance toward homosexuality. Journal of Psychology and Theology, 9(1), 40-57. Campney, B. M. (2015). Race Always Mattered: Black-on-Black Mob Violence and Interracial Relations in Kansas. American Nineteenth Century History, 16(1), 35-57. Cardwell, J. D. (1980). The social context of religiosity. Lanham, MD: University Press of America. Christiano, K. J. (1987). Religious diversity and social change: American cities, 1890-1906. Cambridge University Press. 97 University of Ghana http://ugspace.ug.edu.gh Citifmonline, (2011, April 1). 'Woman thief' stripped naked and molested by University students’. Ghanaweb.com Retrieved from https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Woman-thief-stripped-naked- and-molested-by-University-students-206122 Connelly, L. M. (2008). Pilot studies. Medsurg Nursing, 17(6), 411. Cook, N. (2013). "South Africa: Politics, Economy, and U.S. Relations" Congressional Research Service 1-26 Costa, P. T., & McCrae, R. R. (1992). Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological assessment, 4(1), 5. Creswell, J. W., Plano Clark, V. L., Gutmann, M. L., & Hanson, W. E. (2003). Advanced mixed methods research designs. Handbook of mixed methods in social and behavioral research, 209, 240. Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of management, 31(6), 874-900. Daily Guide (2015, August 19). “Nima Gay Man Attackers Being Chased” Peacefmonline.com, Retrieved from http://www.peacefmonline.com/pages/local/social/201508/251581.php Daily Guide, (2010, Novermber 26). “Witchcraft: Grandma set ablaze”. Ghanaweb.com. Retrieved from http://www.ghanaweb.com/GhanaHomePage/NewsArchive/Witchcraft- Grandma-set-ablaze-198244 De Coster, S., Heimer, K., & Wittrock, S. M. (2006). Neighborhood disadvantage, social capital, street context, and youth violence. The Sociological Quarterly, 47(4), 723-753. 98 University of Ghana http://ugspace.ug.edu.gh Diener, E. (1976). Effects of prior destructive behavior, anonymity, and group presence on deindividuation and aggression. Journal of Personality and Social Psychology, 33(5), 497. Diener, E., Lusk, R., DeFour, D., & Flax, R. (1980). Deindividuation: Effects of group size, density, number of observers, and group member similarity on self-consciousness and disinhibited behavior. Journal of Personality and Social Psychology, 39(3), 449. Donkor, B.K (2015, March 30). “Suspected robbers burnt to death in Kumasi”. Graphic Online. Retrieved from https://www.graphic.com.gh/news/general-news/suspected-robbers-burnt- to-death-in-kumasi.html Dukes R, Clayton S, Jenkins L, Miller T, Rodgers S, (2001). Effects of aggressive driving and driver characteristics on road rage. The Social Science Journal. 38(2):323 -331 Dzokoto, V. A., & Adams, G. (2005). Understanding genital-shrinking epidemics in West Africa: koro, juju, or mass psychogenic illness?. Culture, medicine and psychiatry, 29(1), 53-78. Eisenberger, R., Lynch, P., Aselage, J., & Rohdieck, S. (2004). Who takes the most revenge? Individual differences in negative reciprocity norm endorsement. Personality and Social Psychology Bulletin, 30(6), 787-799. Evans-Pritchard, E. E. (1937). Witchcraft, oracles and magic among the Azande (Vol. 12). London: Oxford. Fanon, F. (1963). Concerning violence. The wretched of the earth, 35-106. Federal Bureau of Investigation, (1935 – 1997). Uniform crime report. Washington, DC: Government printing office. 99 University of Ghana http://ugspace.ug.edu.gh Festinger, L., Pepitone, A., & Newcomb, T. (1952). Some consequences of de-individuation in a group. The Journal of Abnormal and Social Psychology, 47(2S), 382. Fourchard, L. (2008). A new name for an old practice: Vigilantes in south-western Nigeria. Africa, 78(1 1), 16-40. French, P., Anderson, J., Burnard, P., Holmes, C., Mashaba, G., Wong, T., & Bing- hua, Z. (1996). International comparison of baccalaureate nursing degrees: collaboration in qualitative analysis. Journal of Advanced Nursing, 23(3), 594-602. Frey, K. S., Pearson, C. R., & Cohen, D. (2015). Revenge is seductive, if not sweet: Why friends matter for prevention efforts. Journal of Applied Developmental Psychology, 37, 25-35. Gadzekpo, A. (2010). Global warning or global warming? the framing of climate change discourses in Ghanaian newspapers. Ghana Social Science Journal, 7(1), 72-86. Garlipp, P. (2008). Koro–A culture-bound phenomenon intercultural psychiatric implications. German Journal of Psychiatry, 11(1), 21-28. Girard, R. (1977). The Scapegoat. Translated by Yvonne Freccero. London: Athlone, 1986.. Violence and the Sacred. Translated by Patrick Gregory. Glassner, B. and Owen, W. 1976. Variation in attitudes towards homosexuals. Cornell Journal of Social Relations, 11: 161–176. Goldbaum, C., (2014), ‘Kin mourn boy killed in crossfire’, Cape Times, 24 July, 2014, p. 6. Goldstein, D. M. (2003). "In our own hands": Lynching, justice, and the law in Bolivia. American Ethnologist, 30(1), 22–43. Gore, C., & Pratten, D. (2003). The politics of plunder: the rhetorics of order and disorder in 100 University of Ghana http://ugspace.ug.edu.gh southern Nigeria. African Affairs, 102(407), 211-240. Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American sociological review, 161-178. Graneheim, U. H., & Lundman, B. (2004). Qualitative content analysis in nursing research: concepts, procedures and measures to achieve trustworthiness. Nurse education today, 24(2), 105-112. Gray, N. (2016). Mensah Adinkrah. Witchcraft, Witches and Violence in Ghana. New York: Berghahn Books, 2015. xii+ 325 pp. Maps. Photographs and Illustrations. Glossary. Bibliography. Index. $110.00. Cloth. ISBN: 978-1-78238-560-8. African Studies Review, 59(2), 287-287. Greene, E. J., & Darley, J. M. (1998). Effects of necessary, sufficient, and indirect causation on judgments of criminal liability. Law and Human Behavior, 22(4), 429-451. Gyimah-Boadi, E. (2009). Another step forward for Ghana. Journal of Democracy, 20(2), 138- 152. Haas, N. E. (2010). Public support for vigilantism. Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) in collab. with the Leiden Law School, Leiden University. Haas, N. E., de Keijser, J. W., & Bruinsma, G. J. (2012). Public support for vigilantism: an experimental study. Journal of experimental criminology, 8(4), 387-413. Haas, N. E., de Keijser, J. W., & Bruinsma, G. J. (2014). Public support for vigilantism, confidence in police and police responsiveness. Policing and society, 24(2), 224-241. Häefele, B. (2006). Vigilantism in the Western Cape. South Africa: DCS. 101 University of Ghana http://ugspace.ug.edu.gh Hafer, C. L., & Begue, L. (2005). Experimental research on just-world theory: problems, developments, and future challenges. Psychological bulletin, 131(1), 128. Hagen, R., Makovi, K., & Bearman, P. (2013). The influence of political dynamics on southern lynch mob formation and lethality. Social forces, 92(2), 757-787. Hamber, B., & Lewis, S. (1997). An overview of the consequences of violence and trauma in South Africa. Centre for the Study of Violence and Reconciliation. Harnischfeger, J. (2003). The Bakassi boys: fighting crime in Nigeria. The Journal of Modern African Studies, 41(1), 23-49. Hartinger-Saunders, R. M., Rine, C. M., Nochajski, T., & Wieczorek, W. (2012). Neighborhood crime and perception of safety as predictors of victimization and offending among youth: A call for macro-level prevention and intervention models. Children and youth services review, 34(9), 1966-1973. Heimer, K., & De Coster, S. (2001). Crime and gender. Hendricks-Ferguson, V. (2006). Relationships of age and gender to hope and spiritual well-being among adolescents with cancer. Journal of Pediatric Oncology Nursing, 23(4), 189-199. Herek, G. M. (1984). Beyond" homophobia": A social psychological perspective on attitudes toward lesbians and gay men. Journal of homosexuality, 10(1-2), 1-21. Hertzog, M. A. (2008). Considerations in determining sample size for pilot studies. Research in nursing & health, 31(2), 180-191. Higazi, A. (2008). Social mobilization and collective violence: vigilantes and militias in the 102 University of Ghana http://ugspace.ug.edu.gh lowlands of Plateau State, Central Nigeria. Africa, 78(1 1), 107-135. Hofmans, J., Kuppens, P., & Allik, J. (2008). Is short in length short in content? An examination of the domain representation of the Ten Item Personality Inventory scales in Dutch language. Personality and Individual Differences, 45(8), 750-755. Hough, M., Jackson, J., & Bradford, B. (2013). Legitimacy, trust and compliance: An empirical test of procedural justice theory using the European Social Survey. Hough, M., Jackson, J., Bradford, B., Myhill, A., & Quinton, P. (2010). Procedural justice, trust, and institutional legitimacy. Policing: a journal of policy and practice, 4(3), 203-210. Initiative, E. J. (2015). Lynching in America: Confronting the legacy of racial terror (report summary). Inter-university Consortium for Political and Social Research. (2001). “Historical, De- mographic, Economic, and Social Data: The United States, 1790–1970.” Computer file ICPSR00003-v1. ICPSR, Ann Arbor, Mich. doi:10.3886/ICPSR00003. Jackson, J., Asif, M., Bradford, B., & Zakria Zakar, M. (2014). Corruption and police legitimacy in Lahore, Pakistan. British Journal of Criminology, 54(6), 1067-1088. Jacobs, C., & Schuetze, C. (2011). Violence against Violence: In Search of Security and Justice. Etnofoor, 23(2), 17. Jacobs, D., Carmichael, J. T., & Kent, S. L. (2005). Vigilantism, current racial threat, and death sentences. American Sociological Review, 70(4), 656-677. Janis, I. L. (1972). Victims of groupthink: a psychological study of foreign-policy decisions and fiascoes. 103 University of Ghana http://ugspace.ug.edu.gh Johnson, R. D., & Downing, L. L. (1979). Deindividuation and valence of cues: Effects on prosocial and antisocial behavior. Journal of Personality and Social Psychology, 37(9), 1532. Kagwanja, P. M. (2003). Facing Mount Kenya or facing Mecca? The Mungiki, ethnic violence and the politics of the Moi succession in Kenya, 1987–2002. African Affairs, 102(406), 25- 49. Kagwanja, P. M. (2005). ‘Power to Uhuru’: Youth identity and generational politics in Kenya’s 2002 Elections. African Affairs, 105(418), 51-75. Kan, R. W. M., Au, K. P., Chan, W. K., Cheung, L. W. M., Lam, C. Y. Y., Liu, H. H. W., ... & Wong, W. C. (2009). Homophobia in medical students of the University of Hong Kong. Sex Education, 9(1), 65-80. Keil, T. J., & Vito, G. F. (2009). Lynching and the Death Penalty in Kentucky, 1866–1934: Substitution or Supplement?. Journal of Ethnicity in Criminal Justice, 7(1), 53-68. Krejcie, R. V., & Morgan, D. W. (1970). Determining sample size for research activities. Educational and Psychological Measurement, 30, 607-610. Krivo, L. J., & Peterson, R. D. (1996). Extremely disadvantaged neighborhoods and urban crime. Social forces, 75(2), 619-648. Lackey, N. R., & Wingate, A. L. (1997). The Pilot Study: One Key. Advanced design in nursing research, 375. Lamptey, A. A., & Salihu, N. (2012). Interrogating the relationship between the politics of 104 University of Ghana http://ugspace.ug.edu.gh patronage and electoral violence in Ghana. Managing Election-Related Violence for Democratic Stability in Ghana. Accra: Friedrich-Ebert-Stiftung Ghana, 176-209. Le Bon, G. (2017). The crowd. Routledge. Lenţa, O. E. (2014). Violence Dynamics in the context of Social Redefining in a Changing Society. Procedia-Social and Behavioral Sciences, 149, 484-489. Lerner, M. J. (1980). The belief in a just world. In The Belief in a just World (pp. 9-30). Springer, Boston, MA. Lodewijkx, H. F., & Brouwer, B. (2011). Some empirical notes on the epo epidemic in professional cycling. Research quarterly for exercise and sport, 82(4), 740-754. Lodewijkx, H. F., Wildschut, T., Nijstad, B. A., Savenije, W., & Smit, M. (2001). In a violent world a just world makes sense: The case of “senseless violence” in The Netherlands. Social Justice Research, 14(1), 79-94. Lopez V, Emmer E, (2002). Influence of beliefs and values on male adolescents decision to commit violents offenses. Psychology of Men & Masculinity. 3(1):28 - 40 Lumby, M. E. (1976). Homophobia: The quest for a valid scale. Journal of Homosexuality, 2(1), 39-48. Madienyane, D. (2013). Effects of vigilantism on the community of Diesploof. University of Witwatersrand. Available at http://www.wiredspace.wits.ac.za URI: http://hdl.handle.net/10539/13113 Marchal, R. (2007). Warlordism and terrorism: how to obscure an already confusing crisis? The case of Somalia. International Affairs, 83(6), 1091-1106. 105 University of Ghana http://ugspace.ug.edu.gh Martin, J. R. (2010). Vigilantism and informal social control in South Africa. Acta criminologica: Southern African journal of criminology, 23(3), 53-70. McCullough, M. E., Bellah, C. G., Kilpatrick, S. D., & Johnson, J. L. (2001). Vengefulness: Relationships with forgiveness, rumination, well-being, and the Big Five. Personality and social psychology bulletin, 27(5), 601-610. McLean Parks, J. (1998). The fourth arm of justice: The art and science of revenge. Research on Negotiation in Organizations, 6, 113–144. Messner, S. F., Baumer, E. P., & Rosenfeld, R. (2006). Distrust of government, the vigilante tradition, and support for capital punishment. Law & Society Review, 40(3), 559-590. Migliorini, L., Rania, N., Cardinali, P., & Manetti, M. (2008). Sense of safety and the urban environment: A study of preadolescents and adolescents. Medio Ambiente y Comportamiento Humano, 9(1-2), 69-89. Millham, J., San Miguel, C. L., & Kellogg, R. (1976). A factor-analytic conceptualization of attitudes toward male and female homosexuals. Journal of Homosexuality, 2(1), 3-10. Minnaar, A. (2001). The new vigilantism in post-April 1994 South Africa: crime prevention or an expression of lawlessness. Institute for Human Rights and Criminal Justice Studies, May. Mitchell, M. S., & Ambrose, M. L. (2007). Abusive supervision and workplace deviance and the moderating effects of negative reciprocity beliefs. Journal of Applied Psychology, 92(4), 1159. Monaghan, R. (2008). Community-based justice in Northern Ireland and South Africa. International Criminal Justice Review, 18(1), 83-105. 106 University of Ghana http://ugspace.ug.edu.gh Montada, L., & Lerner, M. J. (1998). Critical issues in social justice. Responses to victimizations and belief in a just world. New York, NY, US. Mordey, E.K (2015, May 5). “Two ex-convicts lynched at Jirapa”. Graphic Online. Retrieved fron http://www.graphic.com.gh/news/general-news/two-ex-convicts-lynched-at-jirapa.html Mullen, M. J., Kharbanda, R. K., Cross, J., Donald, A. E., Taylor, M., Vallance, P., & MacAllister, R. J. (2001). Heterogenous nature of flow-mediated dilatation in human conduit arteries in vivo: relevance to endothelial dysfunction in hypercholesterolemia. Circulation research, 88(2), 145-151. Myjoyonline (2017, March 24) “Vandalism: 200 NPP 'Delta Force' attack Ashanti Regional Security Capo”. Myjoyonline.com. Retrieved from https://www.myjoyonline.com/politics/2017/March-24th/ashanti-regional-security- coordinator-attacked.php Newburn, T., & Stanko, E. A. (2013). Just boys doing business?: men, masculinities and crime. Routledge. Ng’walali, P. M., & Kitinya, J. N. (2006). Mob justice in Tanzania: a medico-social problem. African health sciences, 6(1), 36-38. Ngondi-Houghton, C. (2006). Access to Justice and the Rule of Law in Kenya. Commission for the Empowerment of the Poor, Unpublished Manuscript. Quarshie, E.N, Osafo, J., Akotia, C. S., & Peprah, J. (2015). Adolescent suicide in Ghana: A content analysis of media reports. International journal of qualitative studies on health and 107 University of Ghana http://ugspace.ug.edu.gh well-being, 10(1), 27682. Quarshie, E. N., Osafo, J., Akotia, C. S., & Peprah, J. (2015). Adolescent suicide in Ghana: A content analysis of media reports. International journal of qualitative studies on health and well-being, 10(1), 27682. Ninsin, K. A. (Ed.). (1998). Ghana: Transition to democracy. African Books Collective. North, T., & Historical, C. (2016). North Carolina Office of Archives and History Defending White Supremacy: David Clark and the “Southern Textile Bulletin”, 1911 to 1955 Author (s): Bart Dredge Published by: North Carolina Office of Archives and History Stable URL: http://www.jstor.org/stable/23523944 references Linked references are available on JSTOR for this article: Defending White Supremacy: David Clark and the Southern Textile Bulletin, 89(1), 59–91. Nukunya, G. K. (2003). Tradition and change in Ghana: An introduction to sociology. Ghana Universities Press. Nyberg, K. L., & Alston, J. P. (1977). Analysis of public attitudes toward homosexual behavior. Journal of Homosexuality, 2(2), 99-108. Orock, R. T. (2014). Crime, in/security and mob justice: the micropolitics of sovereignty in Cameroon. Social Dynamics, 40(2), 408-428. Osei, B. K. (2011). Role of non-governmental organisations in mitigating gender based violence in Ghana: A case study of Ark Foundation in Accra metropolis (Doctoral dissertation, University of Cape Coast). Osgood, J. M. (2017). Is revenge about retributive justice, deterring harm, or both?. Social and 108 University of Ghana http://ugspace.ug.edu.gh Personality Psychology Compass, 11(1), e12296. Oteng-Ababio, M., Owusu, A. Y., Owusu, G., & Wrigley-Asante, C. (2017). Geographies of crime and collective efficacy in urban Ghana. Territory, Politics, Governance, 5(4), 459- 477. Outwater, A. H., Campbell, J. C., & Mgaya, E. (2011). Implementation of WHO/CDC guidelines for intentional injury death surveillance: A mixed-methods approach in Dar es Salaam, Tanzania. Homicide studies, 15(3), 253-267. Outwater, A. H., Mgaya, E., & Campbell, J. C. (2013). Community violence in Dar es Salaam, Tanzania: A mixed methods study. African Safety Promotion: A Journal of Injury and Violence Prevention, 11(1), 25-38. Pearce-Moses, R., & Baty, L. A. (2005). A glossary of archival and records terminology (Vol. 2013). Chicago, IL: Society of American Archivists. Pedersen, E. J., Kurzban, R., & McCullough, M. E. (2013). Do humans really punish altruistically? A closer look. Proc. R. Soc. B, 280(1758), 20122723. Perugini, M., Gallucci, M., Presaghi, F., & Ercolani, A. P. (2003). The personal norm of reciprocity. European Journal of Personality, 17(4), 251-283. Pfeifer, M. J. (2004). Rough justice: Lynching and American society, 1874-1947. University of Illinois Press. Piper, L., Charman, A., & Carciotto, S. (2016). African human mobility review, 2(1). Polit, D. & Hungler, B. (1999). Nursing Research: Principle and Method, 6th ed.; Philadelphia: Lippincott Company, , P.P. 416-417. 109 University of Ghana http://ugspace.ug.edu.gh Postmes, T., & Spears, R. (1998). Deindividuation and antinormative behavior: A meta-analysis. Psychological bulletin, 123(3), 238. Prah, M., & Yeboah, A. (2011). Tuobodom chieftaincy conflict in Ghana: A review and analysis of media reports. The Journal of Pan African Studies, 4(3), 20-33. Prasad, Rekha (2008, February 26) ‘India’s Pink-clad Vigilantes’, The Guardian. Pratten, D. (2008). ‘The thief eats his shame’: Practice and Power in Nigerian Vigilantism. Africa, 78(1 1), 64-83. Pulse.com.gh (28 September 2016) “In Nungua: One shot dead over land dispute” Pulse online, available online at: https://www.pulse.com.gh/news/in-nungua-one-shot-dead-over-land- dispute-id5541618.html Pulse.com.gh (4 May 2016) “Crime: Nigerian stabbed to death over stolen tooth brush” Pulse online, available online at: https://www.pulse.com.gh/news/crime-nigerian-stabbed-to- death-over-stolen-tooth-brush-id4996703.html Reiss A, Roth J. (1993). Understanding and preventing violence. Vol. 1. Washington, DC, US: National Academy press; p.464 - 470 Richards, P. (1998). Fighting for the rain forest: war, youth & resources in Sierra Leone (No. Reprinted Ed.). James Currey Ltd Rocque, M., Posick, C., & Hoyle, J. (2015). Age and crime. The Encyclopedia of Crime and Punishment, 1-8. Roxburgh, S. (2014). Witchcraft, Violence and Mediation in Africa: A Comparative Study of Ghana and Cameroon (Doctoral dissertation, Université d'Ottawa/University of Ottawa). 110 University of Ghana http://ugspace.ug.edu.gh Salzani, C. (2007). The city as crime scene: Walter Benjamin and the traces of the detective. New German Critique, (100), 165-187. Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social- disorganization theory. American journal of sociology, 94(4), 774-802. Schnoebelen, J. (2009). Witchcraft allegations, refugee protection and human rights: a review of the evidence. UNHCR, Policy Development and Evaluation Service. Sekhonyane, M., & Louw, A. (2002). Violent Justice Vigilantism and the State's Response. Sen, A., Bacchetta, P., Jolly, M., Jensen, S., Nkomazi, P., Africa, S. & Abrahams, R. (2005). Global vigilantes Programme, (July), 10–11. Sharkin, B S. (2004). Road rage: risk factors, assessment, and intervention strategies. J Couns Dev., 82: 191 - 198 Sharpe, J. P., & Desai, S. (2001). The revised Neo Personality Inventory and the MMPI-2 Psychopathology Five in the prediction of aggression. Personality and Individual Differences, 31(4), 505-518. Silke, A. (1998). Motives of paramilitary vigilantism in Northern Ireland. Low Intensity Conflict & Law Enforcement, 7(2), 121-156. Silke, A. (2001). Dealing with vigilantism: issues and lessons for the police. The Police Journal, 74(2), 120-133. Smith, D. J. (2004). The Bakassi boys: vigilantism, violence, and political imagination in Nigeria. Cultural Anthropology, 19(3), 429-455. Smith, D. J. (2010). Response to Eric Uslaner's review of A Culture of Corruption: Everyday 111 University of Ghana http://ugspace.ug.edu.gh Deception and Popular Discontent in Nigeria. Perspectives on Politics, 8(4), 1176-1177. Sommer, M., Likindikoki, S., & Kaaya, S. (2013). Boys' and young men's perspectives on violence in Northern Tanzania. Culture, health & sexuality, 15(6), 695-709. Stavrou, P. (1993). Psychological effects of criminal and political violence on children. Centre for the Study of Violence and Reconciliation. Stavrou, V., 2011, Psychological effects of criminal and political violence on children, Centre for the Study of Violence and Reconciliation, viewed 16 October 2015, from http://www.csvr.org.za Steffensmeier, D., & Allan, E. (1996). Gender and crime: Toward a gendered theory of female offending. Annual review of sociology, 22(1), 459-487. Sundar, N. (2010). Vigilantism, culpability and moral dilemmas. Critique of Anthropology, 30(1), 113-121. Sunshine, J., & Tyler, T. R. (2003). The role of procedural justice and legitimacy in shaping public support for policing. Law & society review, 37(3), 513-548. Super, G. (2016). Volatile sovereignty: Governing crime through the community in Khayelitsha. Law & Society Review, 50(2), 450-483. Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. Allyn & Bacon/Pearson Education. Tankebe, J. (2009). Self- help, policing, and procedural justice: Ghanaian vigilantism and the rule of law. Law & Society Review, 43(2), 245-270. Tankebe, J. (2010). Public confidence in the police: Testing the effects of public experiences of 112 University of Ghana http://ugspace.ug.edu.gh police corruption in Ghana. The British Journal of Criminology, 50(2), 296-319. Tankebe, J. (2013). Viewing things differently: The dimensions of public perceptions of police legitimacy. Criminology, 51(1), 103-135. Tankebe, J., & Asif, M. (2016). Police legitimacy and support for vigilante violence in Pakistan. International Journal of Comparative and Applied Criminal Justice, 40(4), 343-362. Tankebe, J., Reisig, M. D., & Wang, X. (2016). A multidimensional model of police legitimacy: A cross-cultural assessment. Law and Human Behavior, 40(1), 11. Taussig-Rubbo, M. (2011). Pirate Trials, the International Criminal Court, and Mob Justice: Reflections on Postcolonial Sovereignty in Kenya. Humanity: An International Journal of Human Rights, Humanitarianism, and Development, 2(1), 51-74. Tawiah, O. (2017, March 10). “Police arrest man involved in lynching Kumasi taxi driver” Myjoyonline.com. Retrieved from https://www.myjoyonline.com/news/2017/march- 10th/police-arrest-man-involved-in-lynching-kumasi-taxi-driver.php The Herald (31 May 2014) “Tamale Goes Wild Over Arrest Of Gay Men” The Herald online, available online at: http://theheraldghana.com/tamale-goes-wild-arrest-gay-men/ Tolnay, S. E., & Beck, E. M. (1995). A festival of violence: An analysis of southern lynchings, 1882-1930. University of Illinois Press. Tolnay, S. E., & Beck, E. M. (1995). A festival of violence: An analysis of southern lynchings, 1882-1930. University of Illinois Press. Tshehla, M. S. (2014). Africa, where art thou? Pondering post-apartheid South African New 113 University of Ghana http://ugspace.ug.edu.gh Testament scholarship. Neotestamentica, 48(2), 259-281. Tyler, T. R. (2006). Psychological perspectives on legitimacy and legitimation. Annu. Rev. Psychol., 57, 375-400. U.S. Bureau of the Census. (1894). Statistics of Churches: Eleventh Decennial Report of the Census, 1890. Washington, D.C.: Government Printing Office Ukeje, C. (2001). Youths, violence and the collapse of public order in the Niger Delta of Nigeria. Africa Development/Afrique et Développement, 26(1/2), 337-366. UN Office on Drugs and Crime (UNODC). (2010). Victimization Survey in Ghana: Executive Summary. Retrieved February 22, 2012 from http://www.unodc.org/documents/data-and- analysis/dfa/Executive_Summary_ GHANA_final.pdf United Nations Office on Drugs and Crime. (2013). Intentional homicide count and rate per 100,000 population, by country/territory (2000-2012). Available from United Nations Office on Drugs and Crime website: http://www.unodc.org/gsh/en/data.html. Universal Declaration of Human Rights, United Nations Assistance Mission in Afghanistan, & United Nations. Office of the High Commissioner for Human Rights. (2006). Universal declaration of human rights. United Nations Assistance Mission in Afghanistan (UNAMA), Office of the High Commissioner for Human Rights (OHCHR). USDS (United States Department of State) (2002a), 2001 Country Report on Human Rights Practices: Haiti, Bureau of Democracy, Human Rights and Labor. Available at http://www.state.gov/g/drl/rls/hrrpt/2001/ wha/8332pf.htm USDS (United States Department of State) (2002b), 2001 Country Report on Human Rights 114 University of Ghana http://ugspace.ug.edu.gh Practices: Jamaica, Bureau of Democracy, Human Rights and Labor. Available at http://www.state.gov/g/drl/rls/hrrpt/2001/ wha/8325pf.htm USDS (United States Department of State) (2002c), 2001 Country Report on Human Rights Practices: Peru, Bureau of Democracy, Human Rights and Labor. Available at http://www.state.gov/g/drl/rls/hrrpt/2001/wha/ 8263pf.htm USDS (United States Department of State) (2002d), 2001 Country Report on Human Rights Practices: Venezuela, Bureau of Democracy, Human Rights and Labor. Available at http://www.state.gov/g/drl/rls/hrrpt/2001/ wha/8229pf.htm Veiga, J. F., Floyd, S., & Dechant, K. (2001). Towards modelling the effects of national culture on IT implementation and acceptance. Journal of Information technology, 16(3), 145-158. Wamue, G. N. (2001). Revisiting our indigenous shrines through Mungiki. African Affairs, 100(400), 453-467. Watch, H. R. (2010). Paramilitaries’ Heirs: The New Face of Violence in Colombia. Human Rights Watch. Whitfield, L., & Jones, E. (2007). Ghana: The political dimensions of aid dependence. Global Economic Governance Programme (Negotiating Aid: How recipient governments can expand their policy space, 32). Available online at http://www. globaleconomicgovernance. org/sites/geg/files/GEG% 20WP, 20. Whittemore, R., Chase, S. K., & Mandle, C. L. (2001). Validity in qualitative research. Qualitative health research, 11(4), 522-537. Wilson, W. J. (1978). The declining significance of race. Chicago: University of Chicago Press. 115 University of Ghana http://ugspace.ug.edu.gh Wu, Yuning, and Ivan Y. Sun. "Citizen trust in police: The case of China." Police Quarterly 12, no. 2 (2009): 170-191. 116 University of Ghana http://ugspace.ug.edu.gh APPENDICES Appendix 1: Survey Instrument DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF GHANA, LEGON Dear respondent, I am a final year postgraduate student of the University of Ghana, Legon pursuing an MPhil in Social Psychology under the supervision of Prof Charity Akotia and Dr Francis Annor. As part of the requirements for the award of the degree, I am undertaking a study on Attitudes towards vigilantism in Ghana. The researcher seeks to understand the factors that influence people’s attitudes towards vigilantism in Ghana as well the psychological issues in connection to vigilantism. Your participation in this study is entirely voluntary. Any information you provide will be treated with the highest confidentiality and used for academic purposes only. It would be greatly appreciated if you could spend some time to complete this questionnaire. SECTION A: - DEMOGRAPHIC DATA This section asks you to tell us a bit about yourself. Please answer by ticking the appropriate box or writing your response in the space provided 1. Age: _________ 2. Sex: Male [ ] Female[ ] 3. Marital status: Married [ ] Single [ ] 4. Religion: Christian [ ] Muslim [ ] African Traditional Religion [ ] Other [ ] Please specify other: ________________________ 5. Occupation: ________________________ 6. Residence: ________________________ 117 University of Ghana http://ugspace.ug.edu.gh SECTION B The following statements seeks to understand some aspects of your previous experience of crime in Ghana. In general, thick either YES (1) or NO (2) to answer each of the following statements. Yes No 1. Have you ever been a victim of any kind of violence? 1 2 a. Theft / Robbery 1 2 b. Political turbulence 1 2 c. Land dispute 1 2 d. Rape 1 2 e. Mob justice 1 2 2. Have you witness any of these forms of violence before? 1 2 3. Do you know anyone who has experienced violence before? 1 2 4. were the perpetrators punished? 1 2 5. On a scale of 5, how do you rate Criminal activities in your neighbourhood? Not rampant [ ] Somehow rampant [ ] Not sure [ ] Rampant [ ] Very rampant [ ] 6. On a scale of 5, are you satisfied with how the perpetrators were punished? Not satisfied [ ] Somehow satisfied [ ] Not sure [ ] Satisfied [ ] Very satisfied [ ] 118 University of Ghana http://ugspace.ug.edu.gh SECTION C This section reflects your belief in Negative reciprocity norm. Participants are to respond to the statements by expressing their agreement on a 5-point Likert-type scale (1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree). Strongly Disagree Neut Agree Strongly Disagree ral Agree 1. I belief in the saying, “A tooth for a tooth and an eye for an eye”. 1 2 3 4 5 2. If someone dislikes me, I will dislike the person too. 1 2 3 4 5 3. If a person despises me, I will despise them too. 1 2 3 4 5 4. If someone says something nasty to me, I will say something nasty back. 1 2 3 4 5 5. If a person wants to be my enemy, I will treat them like an enemy. 1 2 3 4 5 6. If someone treats me badly, I feel I should treat them even worse. 1 2 3 4 5 7. If someone treats me badly, I will treat that person badly in return. 1 2 3 4 5 8. If someone important to me does something negative to me, I will do something even more 1 2 3 4 5 negative to them. 9. A person who has contempt for me deserves my contempt. 1 2 3 4 5 10. Anyone who treats me like an enemy, deserves mine resentment. 1 2 3 4 5 11. When someone hurts me, I will find a way they won’t know about to get even. 1 2 3 4 5 12. I will not give help to those who treat me badly. 1 2 3 4 5 13. If you distrust me, I will distrust you too. 1 2 3 4 5 119 University of Ghana http://ugspace.ug.edu.gh SECTION D For each of these statements below, please indicate the extent of your agreement or disagreement in order to determine your potential attitude towards vigilante violence. Kindly tick as appropriate; Strongly disagree (1), disagree (2), Neutral (3), agree (4) or Strongly agree (5), for the following statements. Strongly Strongly Disagree Disagree Neutral Agree Agree 1. It is alright for members of the public to beat up crime suspects. 1 2 3 4 5 2. People who kill armed robbers should not be blamed. 1 2 3 4 5 3. It is sometimes OK for people to take the law into their own hands if they feel the 1 2 3 4 5 police are unable to protect them. 4. It is pointless to hand over a suspected criminal to the police because they would not 1 2 3 4 5 bring the offender to justice. 5. I feel satisfied if criminals are punished by community members than the law enforcers. 1 2 3 4 5 6. Beating of thieves is good and should be encourage. 1 2 3 4 5 7. My mood rapidly cycled in terms of anger and anxiety when criminals are caught. 1 2 3 4 5 8. I don’t belief in the right of a criminal. 1 2 3 4 5 9. I believe in punishing those who wrong me or others. 1 2 3 4 5 10. Punishing criminals out of the legal domain is sensible. 1 2 3 4 5 11. Punishing criminals out of the legal domain is disgusting. 1 2 3 4 5 12. The criminal is vulnerable at all time. 1 2 3 4 5 13. Beating of criminals deters criminal activities. 1 2 3 4 5 120 University of Ghana http://ugspace.ug.edu.gh 14. There should be no restriction on beating criminals caught. 1 2 3 4 5 SECTION E This section requires you to tell us about your perception of police trustworthiness. Kindly tick as appropriate; Strongly disagree (1), disagree (2), Neutral (3), agree (4) or Strongly agree (5), with the following statements that reflects how you perceived the police. Strongly Strongly Disagree Neutral Agree Disagree Agree 1.The Ghana police are trustworthy. 1 2 3 4 5 2. I have confidence in the Ghana police. 1 2 3 4 5 3. The Ghana police are usually honest. 1 2 3 4 5 4. The Ghana police always act within the law. 1 2 3 4 5 5. The police can be trusted to make decisions that are right for the people in your 1 2 3 4 5 neighborhood. 6. People’s basic rights are well protected by the police in your neighborhood. 1 2 3 4 5 7. The police in your neighborhood are generally honest. 1 2 3 4 5 8. Ghana has one of the best police forces in Africa. 1 2 3 4 5 9. I am proud of the work of the Ghana police force. 1 2 3 4 5 10. I am happy to defend the work of the Ghana police force when talking to my friends. 1 2 3 4 5 11. I agree with many of the values that define what the Ghana police force stands for. 1 2 3 4 5 12. I cannot think of another police force that I respect more than the Ghana police force. 1 2 3 4 5 13. The work of the Ghana police force encourages me to feel good about our country. 1 2 3 4 5 121 University of Ghana http://ugspace.ug.edu.gh SECTION F For each of the statements below, please indicate the extent of your agreement or disagreement to indicate your sense of safety. Kindly tick as appropriate; Strongly disagree (1); disagree (2); neutral (3); agree (4); strongly agree (5); for the following statements Strongly Disagree Neutral Agree Strongly Disagree Agree 1. Overall, my neighborhood is a good 1 2 3 4 5 place to live in terms of security. 2. There are not many instances of 1 2 3 4 5 crime in my neighborhood. 3. Overall, the police are doing a good 1 2 3 4 5 job in my neighborhood. 4. Crime levels in my neighborhood 1 2 3 4 5 have changed for the better in the last year. 5. I feel unsafe being alone. 1 2 3 4 5 6. To what extent do you feel safe in your neighborhood? Unsafe [ ] Not so safe [ ] Neither [ ] Safe [ ] Very safe [ ] 122 University of Ghana http://ugspace.ug.edu.gh SECTION G The following statements seeks to understand your personality type. In general, do you strongly disagree (1), disagree (2), neutral (3), agree (4), strongly agree (5) with each statement. Strongly Disagree Disagree Neutral Agree Strongly Agree 1. I am a talkative. 1 2 3 4 5 2. I am rather lively. 1 2 3 4 5 3. I enjoy meeting new people. 1 2 3 4 5 4. I usually take the initiative in making new 1 2 3 4 5 friends. 5. I can easily get some life into a rather dull 1 2 3 4 5 place. 6. I tend to be in the background on social 1 2 3 4 5 occasions. 7. I like mixing with people. 1 2 3 4 5 8. I like plenty of movement and excitement 1 2 3 4 5 around me. 9. Other people think of me as being very 1 2 3 4 5 lively. 10. I can get a party or an activity going. 1 2 3 4 5 11. My mood often goes up and down. 1 2 3 4 5 12. I sometimes feel ‘just miserable’ for no 1 2 3 4 5 reason. 13. I feel easily hurt. 1 2 3 4 5 14. I often feel ‘fed-up’ with things or 1 2 3 4 5 situations. 15. I am a nervous person. 1 2 3 4 5 16. I am a worrier. 1 2 3 4 5 17. I don’t worry too long after an 1 2 3 4 5 embarrassing experience. 123 University of Ghana http://ugspace.ug.edu.gh 18. I suffer from ‘determinations’ or ‘will 1 2 3 4 5 power’. 19. I often feel lonely. 1 2 3 4 5 20. Often times, I am troubled about feelings 1 2 3 4 5 of guilt. SECTION H The following statements seeks to understand your religiosity. Kindly tick as appropriate; Every day (1); Most days (2); Some days (3); Once in a while (4); Never or almost never (5) Some Once Every Most days a Never day days while 1. I feel God’s presence. 1 2 3 4 5 2. I experience a connection to all aspects of life. 1 2 3 4 5 3. During worship, or at other times when connecting 1 2 3 4 5 with God, I feel joy which lifts me out of my daily concerns. 4. I find strength in my religion or spirituality. 1 2 3 4 5 5. I find comfort in my religion or spirituality. 1 2 3 4 5 6. I feel deep inner peace or harmony. 1 2 3 4 5 7. I ask for God’s help in the midst of daily activities. 1 2 3 4 5 8. I feel guided by God in the midst of daily activities. 1 2 3 4 5 9. I feel God’s love for me, directly. 1 2 3 4 5 10.I feel God’s love for me, through others. 1 2 3 4 5 11.I am spiritually touched by the beauty of creation. 1 2 3 4 5 12.I feel thankful for my blessings. 1 2 3 4 5 13.I feel a selfless caring for others. 1 2 3 4 5 14.I accept others even when they do things I think are 1 2 3 4 5 wrong. 15. I desire to be closer to God or in union with Him. 1 2 3 4 5 124 University of Ghana http://ugspace.ug.edu.gh 16. In general, how close do you feel to God? Not at all close [ ] Somewhat close [ ] Not sure [ ] Very close [ ] As close as possible [ ] SECTION I Kindly read the passage below carefully and answer the questions that follows; Abbasi is a community in Accra that has witnessed criminal activities over the years. Tang and Frank resides in this community, as well known as ex-convicts. Tang was recently sentenced to nine months in prison for stealing motorbikes and ruminants by the Abbasi Magistrates Court. He however appealed against the sentence and it was converted to a simple fine which he duly paid. Some few days later, he was spotted moving about freely in the town, a development which the people did not understand. This decision did not go down well with the people, who decided to visit instant justice on all suspected criminals and ex-convicts in the town. The youth lynched Tang and Frank after which they burnt the body of Tang. Strongly Strongly disagree Disagree Neutral Agree agree 1. The community members deserves punishment for what they did. 1 2 3 4 5 2. If the community members get punished, they get what they deserve. 1 2 3 4 5 3. The community members should be criminally prosecuted. 1 2 3 4 5 4. The community members cannot be blamed for anything. 1 2 3 4 5 5. The community’s behavior should absolutely not be tolerated. 1 2 3 4 5 6. The authorities should turn a blind eye to the community’s behavior. 1 2 3 4 5 7. what the community members did is justified. 1 2 3 4 5 8. Given the situation, the community’s behavior is appropriate. 1 2 3 4 5 125 University of Ghana http://ugspace.ug.edu.gh 9. The community’s behavior is completely out of proportion. 1 2 3 4 5 10. Thanks to members of the community, at least something is done against crime. 1 2 3 4 5 11. The community members should have looked for another solution. 1 2 3 4 5 12. Behavior like that of the community forms threat to the legal systems. 1 2 3 4 5 13. Frank and Tang have themselves to be blamed for the community’s reaction. 1 2 3 4 5 14. The community’s reaction is understandable. 1 2 3 4 5 15. I pity Frank and Tang who were targeted by the community members. 1 2 3 4 5 16. The community members should have handed Frank and Tang over to the police. 1 2 3 4 5 Thanks for your participation in this survey. 126 University of Ghana http://ugspace.ug.edu.gh Appendix 2: Media articles SNO SOURCES OF ARTICLE LINK TO ARTICLES YEAR Q001 ghheadlines.com http://ghheadlines.com/agency/3news/20170529/44332947/mistaken-identity-army- 2017 officer-lynched-burnt-as-armed-robber Q002 graphiconline.com http://www.graphic.com.gh/news/general-news/suspected-robbers-burnt-to-death-in- 2015 kumasi.html Q003 graphiconline.com http://www.graphic.com.gh/news/general-news/police-arrest-four-for-lynching-42-year- 2016 old-man.html Q004 graphiconline.com http://www.graphic.com.gh/news/general-news/man-lynched-at-kwabenya.html 2014 Q005 adomonline.com http://www.adomonline.com/ghana-news/ghana-news-female-thief-stripped-naked-in- 2017 public/ Q006 graphiconline.com http://www.graphic.com.gh/news/general-news/two-ex-convicts-lynched-at-jirapa.html 2015 Q007 graphiconline.com http://www.graphic.com.gh/news/general-news/two-ex-convicts-lynched-at-jirapa.html 2015 Q008 graphiconline.com http://www.graphic.com.gh/news/general-news/two-ex-convicts-lynched-at-jirapa.html 2015 Q009 graphiconline.com http://www.graphic.com.gh/news/general-news/two-ex-convicts-lynched-at-jirapa.html 2015 Q010 graphiconline.com http://www.graphic.com.gh/news/general-news/five-arrested-for-lynching-man.html 2014 Q011 graphiconline.com https://www.graphic.com.gh/news/general-news/suspected-motorbike-snatcher- 2014 lynched.html Q012 ghanaweb.com http://www.ghanaweb.com/GhanaHomePage/NewsArchive/Witchcraft-Grandma-set- 2010 ablaze-198244 Q013 Citifmonline http://citifmonline.com/2017/06/02/man-accused-of-being-wizard-nearly-lynched/ 2017 Q014 Citifmonline http://citifmonline.com/2017/06/11/63-suspects-rounded-up-over-lynching-of-alleged- 2017 witch/ Q015 adomonline.com www.adomonline.com/ghana-news/hot-audio-man-55-lynched-stealing-gh¢1-50-pesewas/ 2017 Q016 sites.eveyo.com http://sites.eveyo.com/news/264759-suspected-thief-19-escapes-lynching.html 2017 127 University of Ghana http://ugspace.ug.edu.gh Q017 therawafrica.com https://therawafrica.com/suspected-phone-thief-almost-beaten-to-death-unforgiving-mob- 2016 wanted-to-set-him-ablaze/ Q018 myjoyonline.com http://www.myjoyonline.com/news/2017/february-16th/family-of-37-year-old-man-who- 2017 died-in-police-cell-after-mob-attack-wants-justice.php Q019 pinkfmonlinegh.com http://pinkfmonlinegh.com/rise-mob-justice-ghana/ 2017 Q020 modernghana.com https://www.modernghana.com/news/689224/the-social-evil-called-instant-justice.html 2016 Q021 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Police-Administration- 2017 posthumously-promotes-slain-officers-554530 Q022 ghanaweb.com http://www.ghanaweb.com/GhanaHomePage/NewsArchive/Woman-thief-stripped-naked- 2011 and-molested-by-University-students-206122 Q023 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/I-was-almost-lynched-killed- 2017 because-people-mistook-me-for-a-thief-Young-footballer-543410 Q024 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Police-Administration- 2017 posthumously-promotes-slain-officers-554530 Q025 Gbcghana http://www.gbcghana.com/1.1972336 2017 Q026 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Afia-Tenge-nearly-lynched- 2017 for-abominable-dressing-543731 Q027 Citifmonline http://citifmonline.com/2017/04/03/npps-invincible-forces-allegedly-lock-up-nhis-office/ 2017 Q028 myjoyonline.com http://www.myjoyonline.com/politics/2017/April-12th/thuggery-by-invincible-delta- 2017 forces.php Q029 myjoyonline.com http://www.myjoyonline.com/news/2017/April-21st/invincible-forces-storm-zabzugu- 2017 nhia-office-drive-out-staff.php Q030 myjoyonline.com http://www.myjoyonline.com/news/2017/February-21st/hot-video-npps-invincible-forces- 2017 descend-on-police-officer-at-flagstaff-house.php Q031 Adomonline http://adomonline.com/ghana-news/invincible-forces-seize-trotro-station-assault-drivers- 2017 accra/ Q032 Ghanacelebrities http://www.ghanacelebrities.com/2017/01/09/npp-security-group-invincible-forces-taken- 2017 passport-office/ 128 University of Ghana http://ugspace.ug.edu.gh Q033 myjoyonline.com http://www.myjoyonline.com/politics/2017/March-24th/ashanti-regional-security- 2017 coordinator-attacked.php Q034 pulse.com http://www.pulse.com.gh/news/zabzugu-shs-student-attacked-after-witchcraft-allegation- 2017 id6403801.html Q035 Citifmonline http://citifmonline.com/2015/11/06/suspected-armed-robber-burnt-alive-in-asawase/ 2015 Q036 ghananews24.com http://www.ghananews24.com/news/2-suspected-robbers-burnt-to-death-in-kumasi- 2015 ghanaweb Q037 Citifmonline http://citifmonline.com/2015/09/01/another-suspected-robber-burnt-to-death-at-kasoa/ 2015 Q038 Citifmonline http://citifmonline.com/2015/08/24/kasoa-residents-hunt-down-suspected-robber-burn- 2015 him-to-death/ Q039 myjoyonline.com http://www.myjoyonline.com/news/2017/March-10th/police-arrest-man-involved-in- 2017 lynching-kumasi-taxi-driver.php Q040 pulse.com.gh http://www.pulse.com.gh/news/ashanti-region-man-lynched-at-adehyeman- 2017 id6770800.html Q041 modernghana.com https://www.modernghana.com/news/661230/motorbike-thief-lynched.html 2015 Q042 ghananews24.com http://www.ghananews24.com/news/bike-snatcher-lynched 2012 Q043 peaefmoline.com http://www.peacefmonline.com/pages/local/social/201305/164216.php 2013 Q044 peaefmoline.com http://www.peacefmonline.com/pages/politics/politics/201705/314385.php 2017 Q045 peaefmoline.com http://www.peacefmonline.com/pages/local/social/201508/251581.php 2015 Q046 modernghana.com https://www.modernghana.com/news/217632/soldiers-save-robber.html# 2009 Q047 ghanadistricts.com/ghanaweb http://www.ghanadistricts.com/newsarchive/story.php?id=6538&m=7&y=2007 2007 Q048 modernghana.com https://www.modernghana.com/blogs/133658/7/hospital-administrator-murdered-in-cold- 2007 blood.html Q049 modernghana.com https://www.modernghana.com/news/175151/robber-lynched.html 2010 Q050 modernghana.com https://www.modernghana.com/news/291317/thief-lynched.html 2008 Q051 savannah news http://savannahnewsblogspotcom.blogspot.com/2010/09/normal-0-false-false-false-en-us- 2010 x-none_1094.html 129 University of Ghana http://ugspace.ug.edu.gh Q052 modernghana.com https://www.modernghana.com/news/355815/two-robbers-lynched.html 2011 Q053 modernghana.com https://www.modernghana.com/news/355815/two-robbers-lynched.html 2011 Q054 modernghana.com https://www.modernghana.com/news/355815/two-robbers-lynched.html 2011 Q055 modernghana.com https://www.modernghana.com/news/315949/ecobank-robber-lynched-at-ofankor.html 2011 Q056 modernghana.com https://www.modernghana.com/news/388993/suspected-thief-lynched.html 2012 Q057 modernghana.com https://www.modernghana.com/news/387456/two-suspects-lynched-in-the-presence-of- 2012 police-at-essuasu.html Q058 modernghana.com https://www.modernghana.com/news/387456/two-suspects-lynched-in-the-presence-of- 2012 police-at-essuasu.html Q059 modernghana.com https://www.modernghana.com/news/419240/thief-tied-to-tree.html 2012 Q060 Peacefmonline http://www.peacefmonline.com/pages/local/social/201206/117453.php 2012 Q061 savannah news http://savannahnewsblogspotcom.blogspot.com/2012/07/27-year-old-alleged-robber- 2012 lynched-in.html Q062 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Woman-thief-stripped- 2011 naked-and-molested-by-University-students-206122 Q063 newsghana.com.gh https://www.newsghana.com.gh/married-woman-stripped-naked-and-fingered-for- 2012 insulting-chief/ Q064 exposeghana.com http://exposeghana.com/2013/05/two-suspected-armed-robbers-lynched-in-kumasi/ 2013 Q065 Peacefmonline http://www.peacefmonline.com/pages/local/social/201304/161073.php?storyid=100& 2013 Q066 myjoyonline.com http://www.myjoyonline.com/news/2014/may-23rd/two-suspected-goat-thieves- 2014 lynched.php Q067 myjoyonline.com http://www.myjoyonline.com/news/2014/may-23rd/two-suspected-goat-thieves- 2014 lynched.php Q068 Citifmonline http://citifmonline.com/2014/09/04/suspected-robber-lynched-in-sissala-east/ 2014 Q069 Citifmonline http://citifmonline.com/2014/09/04/suspected-robber-lynched-in-sissala-east/ 2014 Q070 Citifmonline http://citifmonline.com/2014/09/04/suspected-robber-lynched-in-sissala-east/ 2014 Q071 newsghana.com.gh https://www.newsghana.com.gh/thieves-lynched-in-cape-coast/ 2014 130 University of Ghana http://ugspace.ug.edu.gh Q072 The Herald http://theheraldghana.com/tamale-goes-wild-arrest-gay-men/ 2014 Q073 The Ghanaian Times http://www.ghanaiantimes.com.gh/2-armed-robbers-lynched-in-kumasi/ 2015 Q074 myjoyonline.com http://www.myjoyonline.com/news/2015/May-5th/lynched-suspected-armed-robber- 2015 denied-burial-by-two-communities.php Q075 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/regional/Alleged-thief-lynched-at-Apowa- 2016 496913 Q076 pinkfmonlinegh.com http://pinkfmonlinegh.com/westen-region-suspected-thief-lynched-at-inchaban-in-the- 2016 shama-district/ Q077 kapital971.com http://kapital971.com/suspected-robber-lynched-in-sunyani/ 2016 Q078 peacefmonline.com http://www.peacefmonline.com/pages/local/crime/201610/294184.php 2016 Q079 pulse.com.gh http://www.pulse.com.gh/news/in-dansoman-suspected-robber-lynched-to-death- 2016 id5684852.html Q080 pulse.com.gh http://www.pulse.com.gh/news/in-nungua-one-shot-dead-over-land-dispute- 2016 id5541618.html?-71542419.1508531257 Q081 pulse.com.gh http://www.pulse.com.gh/news/crime-nigerian-stabbed-to-death-over-stolen-tooth-brush- 2016 id4996703.html?-71542419.1508531257 Q082 modernghana.com https://www.modernghana.com/sports/684451/head-porter-lynched-at-mankessim.html 2016 Q083 Citifmonline http://regional.citifmonline.com/2015/01/21/suspected-armed-robber-lynched-at-goaso/ 2015 Q084 http://233times.com/2015/08/photos-thief-escapes-lynching-in-kumasi/ 2015 Q085 peacefmonline.com http://www.peacefmonline.com/pages/local/social/201512/263220.php 2015 Q086 modernghana.com https://www.modernghana.com/news/649219/1/suspected-robber-escapes-lynching.html 2015 Q087 Citifmonline http://citifmonline.com/2017/06/06/3-arrested-over-lynching-of-witch-at-nabdam/ 2017 Q088 peacefmonline.com http://www.peacefmonline.com/pages/local/social/201105/40964.php 2011 Q089 ghananews24.com http://www.ghananews24.com/news/2-thieves-escape-death-peace-fm-online 2016 Q090 ghananews24.com http://www.ghananews24.com/news/2-thieves-escape-death-peace-fm-online 2016 Q091 newsghana.com https://www.newsghana.com.gh/another-suspected-thief-lynched-in-gomoa-east/ 131 University of Ghana http://ugspace.ug.edu.gh Q092 myjoyonline.com http://www.myjoyonline.com/news/2016/december-22nd/two-robbers-burned-to-death-in- 2016 bole.php Q093 exposeghana.com http://exposeghana.com/2013/09/ashaiman-motorbike-thief-nearly-beaten-death/ 2013 Q094 kessbenfm.com http://kessbenfm.com/fetish-priest-juju-man-burnt-alive/ 2015 Q095 Daily Guide.com http://dailyguideafrica.com/instant-justice-nima/ 2017 Q096 gcbghana.com http://www.gbcghana.com/1.1972336 2017 Q097 gcbghana.com http://www.gbcghana.com/1.1972337 2017 Q098 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Ashaiman-Thieves-Busted- 2011 226234 Q099 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Ashaiman-Thieves-Busted- 2011 226235 Q100 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Church-thief-grabbed- 2013 271104 Q101 dailyguide.com http://dailyguideafrica.com/suspected-robber-lynched-in-sunyani/ 2016 Q102 pulsegh.com http://www.pulse.com.gh/news/instant-justice-angry-residents-burn-armed-robber-alive- 2015 id4100388.html Q103 reportghananews.com http://www.reportghananews.com/instant-justice-two-young-men-burnt-to-ashes-for- 2013 trying-to-snatch-a-handbag/ Q104 modernghana.com https://plus.google.com/+modernghana/posts/YzM8MJMhqTW 2017 Q105 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Suspected-thief-lynched-at- 2015 Anyaa-344083 Q106 graphiconline.com https://www.graphic.com.gh/news/general-news/suspected-thief-19-escapes-lynching.html 2017 Q107 Citifmonline http://citifmonline.com/2016/07/22/suspected-thief-lynched-in-shama/ 2016 Q108 233times.com http://233times.com/2017/06/sekondi-takoradi-mob-gives-thief-soft-drink-amidst-instant- 2017 justice-photos/ Q109 peacefmonline.com http://www.peacefmonline.com/pages/local/social/201403/194041.php 2014 Q110 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Tema-Church-thief-busted- 2016 467803 132 University of Ghana http://ugspace.ug.edu.gh Q111 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Alleged-thief-lynched-at- 2008 Adabraka-140859?channel=D1 Q112 exposeghana.com http://exposeghana.com/2013/04/sakumono-two-suspected-armed-robbers-lynched/ 2013 Q113 exposeghana.com http://exposeghana.com/2013/04/sakumono-two-suspected-armed-robbers-lynched/ 2013 Q114 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/economy/artikel.php?ID=149187 2008 Q115 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Suspected-motorbike-thief- 2017 narrowly-escapes-lynching-548111 Q116 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Suspected-motor-bike- 2017 thieves-nearly-lynched-at-Dzorwulu-577656 Q117 todaygh.com http://www.todaygh.com/motorbike-thief-lynched-in-bolga/ 2015 Q118 dailyguide.com http://dailyguideafrica.com/mobile-phone-thief-lynched-kumasi/ 2017 Q119 Pinkfmonline.com http://pinkfmonlinegh.com/rise-mob-justice-ghana/ 2015 Q120 pulsegh.com http://www.pulse.com.gh/filla/panties-thief-amputee-nearly-lynched-for-stealing-panties- 2015 to-satisfy-his-libido-id3571616.html Q121 thinkghana.com https://think.com.gh/old-woman-accused-of-witchcraft-lynched-by-community-in- 2017 tindongo-upper-east-region/ Q122 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/crime/artikel.php?ID=134451&channel=D1 2007 Q123 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/regional/Two-brothers-lynch-woman-75- 2013 they-accused-of-being-a-witch-259973 Q124 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/The-Mayhem-In-Kumasi- 2005 The-Price-Of-Collective-Negligence-80605 Q125 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Gay-man-Arrested-322719 2014 Q126 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Yet-Another-Lynching- 2007 129583 Q127 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/crime/artikel.php?ID=167492&channel=D1 2009 Q128 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Police-save-student-32060 2003 Q129 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/entertainment/Robber-beaten-close-to- 2012 death-233531 133 University of Ghana http://ugspace.ug.edu.gh Q130 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Phone-snatcher-lynched-at- 2008 Tema-145425 Q131 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Phone-snatcher-lynched-at- 2008 Tema-145425 Q132 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/MP-escapes-lynching-28676 2002 Q133 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/regional/Mob-action-Robbery-suspect- 2016 burnt-410754 Q134 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Man-Lynched-On-False- 2006 Alarm-103166 Q135 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Man-lynched-by-mob-for- 2013 murdering-farmer-277188 Q136 ghanaiantimes.html http://www.ghanaiantimes.com.gh/2-armed-robbers-lynched-in-kumasi/ 2015 Q137 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/3-robbers-lynched-at- 2017 Nadowli-559081 Q138 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/3-robbers-lynched-at- 2017 Nadowli-559081 Q139 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/3-robbers-lynched-at- 2017 Nadowli-559081 Q140 Citifmonline http://citifmonline.com/2017/10/04/20-year-old-suspected-thief-beaten-to-pulp-at- 2017 asawase/ Q141 ghanaiantimes.html https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Two-suspected-goat-thieves- 2014 beaten-to-death-310673 Q142 ghanaiantimes.html https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Two-suspected-goat-thieves- 2014 beaten-to-death-310673 Q143 topnews ghana.htm https://www.topnewsghana.com/suspected-donkey-thief-lynched/ 2017 Q144 ghanaweb.com https://www.ghanaweb.com/GhanaHomePage/NewsArchive/Suspected-armed-robber- 2017 lynched-in-Kasoa-537317 Q145 yen.com.gh https://yen.com.gh/44896-mob-lynches-alleged-thief-in-accra.html#44896 2016 134 University of Ghana http://ugspace.ug.edu.gh Appendix 3: Ethical approval 135