Repeat violent victimisation: evidence review

Averdijk (2011)

Reciprocal effects of victimisation and routine activities

US

• Based on the 1994 National Crime Victimisation Survey (NCVS) - an ongoing nationally representative survey (of non-institutionalised persons of aged 12 and over) which collects data on the frequency, characteristics, and consequences of criminal victimisation in the United States
• Questionnaire focuses on victimisation within the last 6 months prior to the first day of the month of the interview
• 108,208 cases included from the years 1995-1998
• Tests the ‘Once Bitten Twice Shy’ perspective, which argues that victimisation decreases risky routine activities which in turn decreases the risk of future victimisation 

• Analysis restricted to public violence – unclear whether applicable to other types of violence (e.g. in the home)
• Only three indicators of routine activities used 
• Not all respondents participated in all waves – problems with missing data and selective attrition
• Survey does not cover the entire population (e.g. limited to crimes committed against those living in private households aged 12 and over)
• Only asks respondents of experiences during recall period of 6 months
• Dependent on respondents recalling past events 
• Data is dated 
• Findings not necessarily applicable to Scotland 

Buss and Abdu (1995)

Repeat victims of violence in an urban trauma centre

US

• Medical record and survey data collected from victims of urban violence presenting at the emergency department of St. Elizabeth Hospital (Youngstown, Ohio) during a 4 year period 
• Study included two subsamples: a retrospective telephone survey (131 respondents) and a personal survey of victim inpatients (102 respondents)
• Data used to assess extent and nature of RVV, including characteristics of repeat victims 

• Small and unrepresentative sample
• Focus on ‘urban violence’ (assault) - unclear whether findings are applicable to other types of violence 
• Only examines victims of RVV who are injured severely enough to be hospitalised
• Findings are now dated 
• Findings not necessarily applicable to Scotland

Clay-Warner et al. (2016)

Differential Vulnerability: Disentangling the Effects of State Dependence and Population Heterogeneity on Repeat Victimisation 

US

• Uses propensity score matching with longitudinal data from the National Crime Victimisation Survey (NCVS) to explore the reasons why repeat violent and property victimisation occur
• Uses three waves of the NCVS (1998-1999), comprising 27,195 cases 

• Abovementioned limitations with NCVS also applicable to this study (e.g. survey does not cover entire population and recall period limited to 6 months)
• NCVS does not contain measures of offending or self-control, so model cannot fully control for risk heterogeneity 
• Data is dated
• Findings not necessarily applicable to Scotland 

Cooper et al. (2000)

Repeat Victims of Violence: Report of a Large Concurrent Case Control Study

US

• Conducted in trauma centre in Baltimore 
• Case-control study which identified 200 cases and 224 controls during a 16 month period, from 1999-2001
• Cases were persons admitted with a traumatic injury as a result of violent assault who had been previously hospitalised for a similar reason
• Controls were a random selection of eligible age- and sex-matched patients admitted for reasons unrelated to violent injury
• Data used to explore risk factors for RVV

• Small and unrepresentative sample
• Focus on violent assault – unclear whether findings are applicable to other types of violence 
• Only examines victims of RVV who are injured severely enough to be hospitalised
• Findings are now dated
• Findings not necessarily applicable to Scotland

Cuevas et al. (2007)

Juvenile delinquency and victimisation: A theoretical typology

US

• Based on a subsample of respondents to the Developmental Victimisation Survey (DVS), which assessed the victimisation experiences of a nationally representative sample of 2,030 young people in the US between 2002-2003
• Subsample included 1,000 respondents aged between 10-17
• Explores connection between delinquency and victimisation and identifies typology of delinquent-victims
• Focus on violent and property victimisation 

• Survey based on self-report data – susceptible to issues of social desirability
• Underrepresentation of types of more severe delinquency and victimisation in sample
• Telephone survey – underrepresents groups without phone access
• Focus on adolescents – unclear whether findings are applicable to other age groups
• Data now dated
• Findings not necessarily applicable to Scotland

Daigle and Fisher (2008)

The violent and sexual victimisation of college women: Is repeat victimisation a problem?

US

• Based on two national-level surveys of more than 8,000 female students: the National College Women Violent Victimisation (NCWVV) and the National College Women Sexual Victimisation (NCWSV)
• Surveys conducted in 1997
• The NCWVV survey measured the extent and nature of different types of violence college women had experienced, whereas the NCWSV survey measured different types of sexual victimisation
• Daigle and Fisher (2008) use this data to explore the extent and nature of both types of victimisation  

• Uses cross-sectional rather than longitudinal design to retrospectively examine violent and sexual victimisation 
• More precise information on date/time of incident required to fully understand the time course of repeat incidents
• Sample of female students – unclear whether findings are applicable to other population groups
• Data now dated
• Findings not necessarily applicable to Scotland

Davis et al. (2006)

Preventing repeat incidents of family violence: analysis of data from three field experiments

US

• Three field experiments conducted in New York during the 1990s to evaluate whether programs, targeted at public housing residents who reported family violence to the police, reduced the rate of subsequent victimisation
• Intervention involved a ‘crisis response team’, consisting of a police officer and social worker, who followed up on initial police response to complaint
• Utilises a sample of nearly 1,000 cases pooled from three randomised experiments

• Small and unrepresentative sample
• Composition of the samples in the three experiments varied, with different populations and offences, preventing valid comparison 
• Focus on family violence – unclear whether applicable to other types of violence
• Data now dated
• Findings not necessarily applicable to Scotland

Deadman and MacDonald (2004) 

Offenders as Victims of Crime? An Investigation into the Relationship Between Criminal Behaviour and Victimisation

England and Wales

• Uses data from the 1998 Youth Lifestyles Survey (YLS)
• Nationally representative sample of 4,848 12-30 year olds living in private households in England and Wales
• Information on offending behaviour is collected through self-completion questionnaires
• Study considers the impact of violent, non-violent and persistent offending on the probability of being a victim of violent and non-violent crime 

• Survey does not cover the entire population (e.g. limited to crimes committed against those living in private households aged 12-30)
• Unclear whether findings are applicable to other age groups
• Only asks respondents of experiences during recall period of 12 months
• Dependent on respondents recalling past events 
• Data is dated 
• Findings not necessarily applicable to Scotland 

Farrell and Pease (1993)

Once bitten, twice bitten: Repeat victimisation and its implication for crime prevention

England and Wales

• Summarises and reviews existing evidence on repeat victimisation
• Covers racial attacks, domestic violence as well as burglary and property crime
• 33 studies are reviewed (conducted between 1967-1993), international focus
• Discusses methodological issues in the study of repeat victimisation

• No systematic assessment of quality of research included
• Findings are now dated
• No Scottish-specific evidence included – findings are not necessarily applicable to Scotland

Farrell (2005)

Progress and prospects in the prevention of repeat victimisation

England

• Summarises and reviews existing evidence relating to the prevention of repeat victimisation 
• Includes 53 studies from the UK, US and elsewhere, conducted between 1974-2005
• Covers various types of repeat victimisation, including personal and violent crimes

• No systematic assessment of quality of research included 
• Findings are now dated 
• No Scottish-specific evidence included – findings are not necessarily applicable to Scotland

Farrell et al. (2005), Kesteren et al. (2013) 

International Crime Victims Survey

41 countries, including Scotland 

• Large-scale, representative international survey conducted across 6 sweeps (1989-1992-1996-2000-2004/05-2010)
• Respondents asked about their experiences of crime over a 5-year period, and then asked to focus on their experiences over the past 12 months
• Can examine the proportion of victims who experienced violence more than once over the course of a year (i.e. the RVV rate)
• Present database covers 325,454 individual respondents

• National samples are relatively small (typically between one and two thousand per country)
• Response rates were variable across countries, reducing comparability
• Samples of repeat victims are small
• Does not cover the entire population (e.g. limited to crimes committed against adults living in private households)
• Cap on number of criminal incidents that person can report as a series
• Only asks questions about recall period of one year
• Dependent on respondents recalling past events 

Farrell and Zimmerman (2017) 

Does offending intensify as exposure to violence aggregates? Reconsidering the effects of repeat victimisation, types of exposure to violence and polyvictimisation on property crime, violent offending and substance use 

US

• Based on two waves of data from the nationally representative National Longitudinal Study of Adolescent to Adult Health (Add Health)
• Sample of 12,603 respondents aged 12-18, collected between 1995-2008
• Data used to explore the relationship between the effects of repeat victimisation, exposure to different types of violence on property crime, violent offending and substance use

• Measure of exposure to violence captured experiences in 12 months preceding interview rather than lifetime exposure
• Focused on adolescents who are enrolled in school; unclear whether findings are applicable to other population groups
• Study is not focused principally on violence
• Data is now dated
• Findings are not necessarily applicable to Scotland

Gerrell (2018)

Bus Stops and Violence, Are Risky Places Really Risky?

Sweden 

• Uses risk terrain modelling (RTM) to geographically forecast risk of crime in Malmo, Sweden
• Compares the spatial risk factors for where the number of crimes is high with the spatial risk factors for where the risk of victimisation is high 
• Performed by fitting negative binomial models on crime around bus stops

• Focuses on robbery and aggravated assault – unclear whether applicable to other types of violence 
• Findings are not necessarily applicable to Scotland

Graham-Kevan et al. (2015)

Repeat Victimisation, Retraumatisation and Victim Vulnerability

England

• Based on qualitative interviews with 54 victims of police recorded violent crime
• Participants recruited from a larger concurrent study into repeat victimisation in Preston
• Data collected from the police database held at Lancashire Constabulary, Victim Support (VS) and Preston Domestic Violence Services (PDVS) between April 2013 and September 2013
• Explores relationship between traumatic experiences, violent crime victimisation and re-victimisation suffered throughout life, in relation to current psychological and psychosocial functioning 

• Limitations with qualitative research include lack of generalisability due to small sample sizes and subjective nature of the research 
• Relies on participant recall of past trauma exposure; may lack accuracy
• Victims self-selected (i.e. responded to requests and attended interviews) – potential for bias
• Based on self-reported symptoms as opposed to clinical assessment 
• Findings are not necessarily applicable to Scotland 

Grove et al. (2012)

Preventing Repeat Victimisation: A Systematic Review

Sweden

• Systematic review of 31 studies evaluating efforts to prevent repeat victimisation
• Evaluation studies selected from systematic searches of databases, hand searches of bibliographies and contact with academics and practitioners 
• Applies inclusion and exclusion criteria and systematic coding manuals
• Crimes types included residential burglary, domestic violence, commercial crime and sexual victimisation

• Review is not focused specifically on violence – unclear whether all findings are applicable to violent crime
• No Scottish-specific evidence included – findings are not necessarily applicable to Scotland

Hanmer et al. (1999)

Arresting Evidence: Domestic Violence and Repeat Victimisation

England 

• Evaluation of a project in Leeds designed to reduce repeat victimisation of domestic violence through the use of early intervention and graduated police responses
• Data collected on extent of repeats from police partner abuse database 
• Quantitative and qualitative research methods used to evaluate the project, including:
• Three-month police recorded partner abuse sample from 1996 (assigned hypothetical intervention levels) compared and tested for statistical significance with three-month period of 1997 (the project period)
• Views and experiences of police officers, agencies and women collected through interviews and questionnaires

• Focus on domestic violence – unclear whether findings applicable to other types of violence
• Data on extent of domestic violence are gathered from police recorded data – limitations of this include underreporting to police and difficulty identifying crimes as repetitions
• Limitations with qualitative interviews include lack of generalisability due to small sample sizes and subjective nature of the research 
• Findings are now dated 
• Findings not necessarily applicable to Scotland

Ignatans and Pease (2015)

On whom does the burden of crime fall now? Changes over time in counts and concentration

England and Wales

• Extracts data from almost 600,000 respondents from the 1982-2012 sweeps of the Crime Survey for England and Wales (CSEW) to determine which types of victimisation have become more or less concentrated during the overall crime drop
• The CSEW is a large-scale, representative survey involving an annual sample of 35,000 households across England and Wales
• Respondents asked about their experiences and perceptions of crime, including crime not reported to the police
• Iganatans and Pease (2015) examine three general crime types (vehicle, property and personal) 

• Limitations of the CSEW include that the survey does not cover the entire population (e.g. limited to adults living in private households)
• Artificial limit of 5 on number of criminal incidents that person can report as a series (lifted from 2018 onwards)
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Findings not necessarily applicable to Scotland

Jacobs and Wright (2006)

Street justice: Retaliation in the criminal underworld

US

• Qualitative interviews with 52 African-American street offenders from St. Louis, Mississippi, who had been the victim of at least one violent crime which they either retaliated or attempted to do so 
• Addresses how criminals seek to avenge themselves after being robbed
• Presents ‘typology of retaliation’ and policy recommendations to address criminal retaliation  

• Limitations with qualitative research include lack of generalisability due to small sample sizes and subjective nature of the research 
• Data now dated 
• Findings not necessarily applicable to Scotland 

Jansson et al. (2007)

Attitudes, perceptions and risks of crime: Supplementary Volume 1 to Crime in England and Wales 2006/07

England and Wales

• Based on the 2006/07 and 2005/06 sweeps of the British Crime Survey (now known as the Crime Survey for England and Wales) – a large, nationally representative survey conducted since 1982
• Asks people about their experiences of property and personal crimes as well as their perceptions of and attitudes towards different crime-related issues
• The 2006/07 BCS included 47,203 face-to-face interviews conducted between April 2006 and March 2007 (47,796 in 2005/06 BCS)
• Jansson et al. (2007) present additional analysis on attitudes, perceptions and risks of crime

• Limitations of the BCS include that the survey does not cover the entire population (e.g. limited to adults living in private households)
• Artificial limit of 5 on number of criminal incidents that person can report as a series 
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Findings not necessarily applicable to Scotland

Jennings et al. (2012)

On the overlap between victimisation and offending: A review of the literature

US

• Review of research examining the relationship between victimisation and offending
• Identifies 37 studies (from 1958-2011) which assess the victim-offender overlap
• Evidence identified uses a range of analytical and statistical techniques 
• Research predominantly from the US but also England and Wales, Canada, South Korea, Sweden and the Netherlands 

• No systematic assessment of quality of research included 
• Some studies included are now dated
• No Scottish-specific evidence included – findings are not necessarily applicable to Scotland

Lauritsen and Davis Quinet (1995)

Repeat victimisation among adolescents and young adults

US

• Uses longitudinal panel data from the nationally representative National Youth Survey (NYS) to test theories of why repeat victimisation occurs  
• Based on a sample of adolescents and young adults
• Interviews conducted from 1977-1982
• Focuses on four crime types – assault, robbery, theft and vandalism 

• Theoretically relevant variables unavailable in NYS data 
• Based on sample of adolescents and young adults – unclear whether findings applicable to other age groups 
• Focus on assault and robbery – unclear whether applicable to other types of violence
• Findings now dated 
• Findings not necessarily applicable to Scotland 

Lloyd et al. (1994)

Preventing Repeated Domestic Violence: A Demonstration Project on Merseyside

England

• Evaluation of the effectiveness of ‘The Merseyside Domestic Violence Prevention Project’, which focused on reducing repeated domestic violence
• Data collected on extent of repeated domestic violence using police incident logs (calls to the police from the public) in Merseyside
• Strategy assessed using qualitative interviews with victims as well as examination of associated costs and the effect on enforceable injunctions

• The design of the study prevented evaluation of the effectiveness in terms of an increase or reduction in repeat victimisation due to the lack of control group 
• Focus on domestic violence – unclear whether findings applicable to other types of violence
• Data on extent of domestic violence are gathered from police recorded data – limitations of this include underreporting to police and difficulty identifying crimes as repetitions
• Limitations with qualitative interviews include lack of generalisability due to small sample sizes and subjective nature of the research 
• Findings are now dated 
• Findings not necessarily directly applicable to Scotland

Matthews et al. (2001)

Repeated Bank Robbery: Themes and Variations

England

• Examines all bank robberies, completed and attempted, reported to the UK Metropolitan Police between 1992-1994
• Demonstrates rate of repetition against the same branches, as well as nature and time course of repeats

• Authors noted difficulties with identifying which robberies were repeats due to police recording practices 
• Focuses specifically on robbery, not necessarily applicable to other types of violence 
• Findings are now dated 
• Findings not necessarily directly applicable to Scotland

min Park and Eck (2013)

Understanding the Random Effect of Victimisation Distributions: A Statistical Analysis of Random Repeat Victimisations

England and Wales

• Simulates data from the 2008/09 British Crime Survey (now known as the Crime Survey for England and Wales) – a large, nationally representative survey conducted since 1982
• Employs mathematical demonstrations to investigate the statistical characteristics of random repeat victimisation
• Aims to clarify the effect of random repeats on the distribution of victimisations 
• The 2008/09 BCS included 46,286 face-to-face interviews conducted between April 2008 and March 2009

• Unclear if ‘random repeat’ pattern identified is applicable to all types of violent crime – particularly crimes demanding special relationships between offenders and victims
• Does not investigate random repeats in the process of multiple victimisation 
• Limitations of the BCS include that the survey does not cover the entire population (e.g. limited to adults living in private households)
• Artificial limit of 5 on number of criminal incidents that person can report as a series 
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Findings not necessarily applicable to Scotland

Morenoff et al. (2001)

Neighbourhood Inequality, Collective Efficacy and the Spatial Dynamics of Urban Violence

US

• Study links police records, statistics, census data and a survey of 8,872 Chicago residents to assess an integrated theoretical perspective on neighbourhood-level variations in homicide 
• Structural characteristics in 1990 and survey measures from 1995 are used to model variations in the event rate of homicide for 1996-1998 across 343 neighbourhoods
• Assesses importance of spatial dynamics for explaining urban violence

• Measures of organisations and institutions drawn from survey (self) reports – limited to residents’ perceptions
• Focus on homicide – unclear whether findings applicable to other types of violence
• Findings are now dated
• Findings not necessarily applicable to Scotland 

Mukherjee and Carach (1993)

Repeat victimisation in Australia: Extent, Correlates and Implications for Crime Prevention

Australia

• Based on data from the 1993 National Crime and Safety Survey Australia (NCSSA) – a large-scale, nationally representative survey – and the 1991 Queensland Crime Victims Survey (Queensland CVS) 
• The NCSSA collects information from individuals and households about their experience of selected crimes, reporting behaviour and risk factors, while the Queensland CVS collects complementary information on fear of crime 
• Surveys include residents of private households
• The NCSSA collects information from those aged 15+, while the Queensland CVS collects information from those aged 18+ 

• Does not cover the entire population (e.g. limited to adults living in private households) 
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Findings now dated
• Findings not necessarily directly applicable to Scotland

Scottish Government (2019)

Scottish Crime and Justice

Survey 2017/18: Partner

Abuse

Scotland

• Analysis based on self-completion module of the 2017/18 sweep of the Scottish Crime and Justice Survey (SCJS) – a large-scale, representative survey asked of 6,000 adults in Scotland each year
• Respondents asked about their experiences of partner abuse within the last month
• Includes incidents not reported to the police

• The SCJS does not cover the entire population (e.g. limited to crimes committed against adults living in private households)
• Dependent on respondents recalling past events

Office for National Statistics (ONS) (2016)

Crime Survey for England and Wales - Intimate personal violence and partner abuse

England and Wales

• Analysis of findings from the self-completion module of the 2014/15 sweep of the Crime Survey for England and Wales – a large scale, nationally representative survey involving a sample of 35,000 households across England and Wales
• Findings on the extent of, and trends in, intimate violence among men and women aged 16 to 59 resident in England and Wales 
• Includes incidents not reported to the police  

• The CSEW does not cover the entire population (e.g. limited to crimes committed against adults living in private households) 
• Dependent on respondents recalling past events
• Findings not necessarily applicable to Scotland 

Office for National Statistics (ONS) (2019)

Crime Survey for England and Wales 

England and Wales

• Large-scale, representative survey involving a sample of 35,000 households across England and Wales
• Respondents asked about their experiences and perceptions of crime
• Includes crimes that have not been reported to the police
• Can examine the proportion of victims who experienced violence more than once over the course of a year (i.e. the RVV rate)
• RVV broken down by types of violence and relationship with offender

• The CSEW does not cover the entire population (e.g. limited to crimes committed against adults living in private households)
• In main survey, artificial limit of 5 on number of criminal incidents that person can report as a series (lifted from 2018 onwards)
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Findings not necessarily directly applicable to Scotland

Ousey et al. (2008)

Déjà vu All Over Again: Investigating Temporal Continuity of Adolescent Victimisation

US

• Uses longitudinal panel data from the Rural Substance Abuse and Violence Project (RSVP) – a study of substance use, criminal victimisation and criminal offending
• Based on a sample of 2,706 adolescents from a public school in Kentucky 
• Data collected between 2001-2004
• Assesses divergent perspectives regarding the state dependence explanation of RVV

• Focus on adolescents (public school students) – unclear whether findings applicable to other population groups 
• Focus on assault and theft victimisation – unclear whether findings applicable to other types of violence 
• Data now dated
• Findings not necessarily applicable to Scotland

Outlaw et al. (2002)

Repeat and Multiple Victimisations: The Role of Individual and Contextual Factors

US

• Based on telephone survey data from a multi-stage sample of 5,302 Seattle residents in 1990
• Uses hierarchical modelling to examine the relative contributions of factors about the person and context in models of repeat victimisation and multiple victimisation
• Estimates separate hierarchical models for repeat property, repeat violent and multiple victimisation 

• Sample underrepresents adults in 20-24 age group and low income families, and over-represents homeowners and married persons
• Only asks respondents of experiences during recall period of two years
• Dependent on respondents recalling past events
• Findings not necessarily applicable to Scotland 

Sampson and Phillips (1992)

Multiple Victimisation: Racial Attacks on an East London Estate 

England 

• Evaluation of a crime prevention project to reduce repeat racial victimisation between 1990 and 1991 on a local authority housing estate in East London 
• Quantitative and qualitative research methods used to evaluate the project, including:
• Collection of statistical data on the number, location and type of racial attacks and harassment from police, local authority housing and community group information
• Qualitative interviews with repeat racial victims, local authority staff, police officers and community group workers and representatives 
• Systematic observations at public meetings

• Based on a small and unrepresentative sample
• The author notes that prevention measures were implemented inconsistently across sample
• Data on extent of racial attacks collected from police - limitations of this include underreporting to police and difficulty identifying crimes as repetitions
• Limitations with qualitative interviews include lack of generalisability and subjective nature of the research
• Focus on racial attacks – unclear whether findings applicable to other types of violence 
• Findings are now dated
• Findings not necessarily applicable to Scotland 

Scottish Government (2019)

Scottish Crime and Justice Survey (SCJS) 2017-18

Scotland

• Large-scale, representative survey asked of 6,000 adults in Scotland each year
• Respondents asked about their experiences and perceptions of crime
• Includes crimes not reported to the police
• Can examine the proportion of victims who experienced violence more than once over the course of a year (i.e. the RVV rate)

• Does not cover the entire population (e.g. limited to crimes committed against adults living in private households)
• In main survey, cap of 5 on number of criminal incidents that person can report as a series
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Small sample sizes for repeat victims 

Schreck et al. (2006) 

Self-control, victimisation and their influence on risky lifestyles: A longitudinal analysis using panel data

US

• Uses the first three waves of panel data from the national evaluation of the Gang Resistance Education and Training (GREAT) program (1995-1999), covering 6 US cities 
• Longitudinal examination of the causal mechanisms underlying violent victimisation using sample of 1,500 adolescents
• Focus on whether victims alter lifestyle choices (like their own delinquency and contact with delinquent peers) in response to their earlier violent victimisation

• Sample attrition and missing data led to reduced sample size
• Only two measures of risky activities included; not covering range of daily activities that could be predictive to victimisation
• Dependent on respondents recalling past events
• Data is now dated
• Findings not necessarily applicable to Scotland 

Shaw and Pease (2000)

Research on Repeat Victimisation in Scotland

Scotland

• Analysis of extent and nature of repeat victimisation in Scotland 
• Triangulation of three methods, including:
• Data from four sweeps of Scottish Crime Survey (SCS) (1982, 1988, 1992, 1996)
• Police recorded crime data for three police divisions, collected over one-year period
• Qualitative interviews with repeat victims and offenders 
• Covers range of crimes, including repeat property victimisation, repeated crime against the person, repeated theft

• Analysis of SCS:
• SCS does not cover the entire population
• Cap on number of criminal incidents that person can report as a series
• Only asks respondents of experiences during recall period of one year
• Dependent on respondents recalling past events
• Police recorded crime:
• Only collected in three police divisions 
• Not all incidents of crime are reported to the police
• Difficult to identify crime as repetition of earlier crime due to recording practices
• Qualitative interviews:
• Limitations with qualitative interviews include lack of generalisability due to small sample sizes and subjective nature of the research 
• Research is not focused specifically, nor principally, on violence 
• Research is now dated

Taylor (2004)

Petrol Service Stations as Victims of Crime: Their Risks and Vulnerabilities 

Australia

• Uses national statistics data from Australian Bureau of Statistics (ABS) on crimes recorded by the Police in each state and territory within Australia from 1993 to 2001, focusing specifically on petrol station robbery
• ABS data combined with national self-report survey data from Small Business Crime Survey (sample of 735 petrol station proprietors)  
• Nature and extent of robberies at petrol stations examined, including levels of repeat victimisation 

• Limitations with police recorded crime include underreporting to police and difficulty identifying crimes as repetitions
• Survey based on self-report data 
• Focuses specifically on robbery - not necessarily applicable to other types of violence 
• Findings are now dated 
• Findings not necessarily applicable to Scotland

Tillyer (2013)

Violent victimisation across the life course: Moving a ‘victims career’ agenda forward

US

• Based on four waves of data from the nationally representative National Longitudinal Study of Adolescent Health (Add Health)
• Sample of 2,779 respondents collected between 1994-2008, from when respondents were aged 12-32
• Examines violent victimisation patterns across the life course and outlines a ‘victims career’ agenda

• Sample attrition and missing data led to reduced sample size (which disproportionately affects high risk individuals)
• Only asks respondents of experiences during recall period of one year - victimisations that occurred outside these time frames are not recorded
• Add Health only includes data through young adulthood - unclear if early and persistent victimisation in adolescence is associated with an increased risk for violent victimisation throughout the life course
• Data is now dated
• Findings are not necessarily applicable to Scotland

Tseloni and Pease (2003)

Repeat personal victimisation: “Flags” or “boosts”?

US

• Based on the 1994 National Crime Victimisation Survey (NCVS) - an ongoing nationally representative survey (of non-institutionalised persons of aged 12 and over) which collects data on the frequency, characteristics, and consequences of criminal victimisation in the United States
• Uses data collected across three waves of interviews between 1994-1995
• Examines effects of state dependence and population heterogeneity on repeat personal victimisations
• Personal victimisations include rape, sexual assault, robbery, assault, threats, pick-pocketing and larceny
• Employs a fixed effects random intercept multilevel model of personal crime counts

• Not all respondents participated in all waves – problems with missing data and selective attrition
• Survey does not cover the entire population (e.g. limited to crimes committed against those living in private households aged 12 and over)
• Only asks respondents of experiences during recall period of 6 months
• Dependent on respondents recalling past events 
• Not focused exclusively on violence 
• Findings are now dated
• Findings not necessarily applicable to Scotland 

Tseloni and Pease (2004)

Repeat Personal Victimisation: Random Effects, Event Dependence and Unexplained Heterogeneity

US

• Based on the 1994 National Crime Victimisation Survey (NCVS) - an ongoing nationally representative survey (of non-institutionalised persons of aged 12 and over) which collects data on the frequency, characteristics, and consequences of criminal victimisation in the United States
• Uses data collected across three waves of interviews between 1994-1995
• Personal victimisations include rape, sexual assault, robbery, assault, threats, pick-pocketing and larceny
• Identifies the risks of repeat personal victimisation and investigates extent to which event dependence and unexplained heterogeneity affect current victimisation 

• Not all respondents participated in all waves – problems with missing data and selective attrition
• Survey does not cover the entire population (e.g. limited to crimes committed against those living in private households aged 12 and over)
• Only asks respondents of experiences during recall period of 6 months
• Dependent on respondents recalling past events 
• Not focused exclusively on violence 
• Findings are now dated
• Findings not necessarily applicable to Scotland

van Reemst et al. (2013)

Risk factors for repeat victimisation: a literature scan

The Netherlands

• Identifies risk factors for repeat victimisation through systematic literature review of 106 studies 
• Search engines used include the search engine of the Erasmus University Rotterdam (sEURch), of the Dutch government, PsychInfo and National Criminal Justice Reference Service
• Studies also searched through snowball and citation methods
• Inclusion and exclusion criteria relating to methods, content and language applied (only prospective studies with strong research design included)

• No primary research conducted
• Possible that not all characteristics and populations have been fully explored in research, or have only been researched in a single study
• Findings are discussed in relation to Dutch people; unclear whether applicable to Scotland

Vecchio (2013)

Once bitten, thrice wise: The varying effects of victimisation on routine activities and risk management

US

• Qualitative study of 36 men (34 African American, 2 white), aged between 19-50, who were participating in a mandatory drug treatment centre in the Midwest
• Interviews conducted in 2010
• Data used to explore effects of victimisation on subsequent behaviour and management of lifestyle risks 

• Limitations with qualitative research include lack of generalisability due to small sample sizes and subjective nature of the research 
• Specifically, study draws on individual narrative accounts of whether victimisations were perceived to subsequently influence behaviour – perceptions are inherently subjective
• Sample is from the US, urban, all male and African American – findings unlikely to be applicable to other population groups

Weisel (2005) 

Analysing Repeat Victimisation

US

• ‘Problem-Oriented Guide for Police’ which summarises and reviews existing knowledge on repeat victimisation 
• 42 published studies reviewed, international focus 
• Covers literature on range of offences, including violence as well as property crime 

• No systematic assessment of quality of research included
• Does not focus exclusively on RVV 
• Findings are now dated
• No Scottish-specific evidence included – Findings unlikely to be applicable to Scotland

Wilcox et al. (2006)

Student weapon possessing and the ‘fear victimisation’ hypothesis

US

• Uses longitudinal data from the Rural Substance Abuse and Violence Project (RSVP) – a study of substance use, criminal victimisation and criminal offending 
• Based on a sample of around 4,000 adolescents from 113 public schools in Kentucky
• Data collected between 2001-2004
• Explores the direction of relationships between student weapon carrying and school-crime experiences, including victimisation 

• Focus on adolescents (public school students) – unclear whether findings applicable to other population groups 
• Substantial number of eligible participants excluded due to non-consent – possible that sample is biased with regard to key variables (e.g. delinquency and victimisation)
• Data is now dated
• Findings not necessarily applicable to Scotland