A Mixed Methods Study into Bystander Experiences of Domestic Violence and Abuse during the COVID-19 Pandemic

Abstract

This research explores the experiences and behaviours of bystanders to domestic abuse during the COVID-19 pandemic, to help inform policy and bystander intervention training programmes in light of the changes to daily lives brought about by COVID-19 restrictions.

This study, delivered in partnership with the University of Exeter, and funded by Public Health Wales, included survey and interviews with members of the public to better understand the experiences and behaviours of bystanders to domestic violence and abuse during a period of social restrictions. Whilst implemented on a small scale, this study was the first of its kind and provides new insights into bystander experiences during a global pandemic.

Full text

Domestic
Violence and
Abuse during
the COVID-19
Pandemic
Bystander Experiences of

A Mixed Methods Study into Bystander Experiences of Domestic
Violence and Abuse during the COVID-19 Pandemic
Research Team
Dr Alex Walker1, Dr Rachel A Fenton2, Lara C Snowdon1, Bryony Parry1, Emma R
Barton1, Prof Catherine Donovan3 and Prof Mark A Bellis4
1 Wales Violence Prevention Unit,
2 School of Law, University of Exeter,
3 Department of Sociology, Durham University,
4 College of Human Sciences, Bangor University.
[October 2021]
ISBN: 978-1-78986-154-455
© 2021 Public Health Wales NHS Trust.
Material contained in this document may be reproduced under the terms of the Open
Government Licence (OGL) www.nationalarchives.gov.uk/doc/open-government-licence/
version/3/ provided it is done so accurately and is not used in a misleading context.
Acknowledgement to Public Health Wales NHS Trust to be stated. Unless stated
otherwise, copyright in the typographical arrangement, design and layout belongs to
Public Health Wales NHS Trust.
This study was funded by Public Health Wales.
This study was delivered by the Wales Violence Prevention Unit
and University of Exeter.
For correspondence related to this report, please contact
Lara.Snowdon@wales.nhs.uk
Wales Violence Prevention Unit
Market Chambers
5-7 St Mary’s Street
Cardiff
CF10 1AT

Acknowledgements 5
Acronyms 6
Executive Summary 7
1.0 Introduction 11
1.1 Background of Domestic Violence and Abuse in Wales 11
  1.2Denition         11
1.3 The Policy Context in Wales 12
1.4 Covid-19 and Domestic Violence and Abuse in Wales 13
1.5 Bystanders to Domestic Violence and Abuse 14
1.6 Study Rationale and Aims 15
2.0 Literature Review 16
2.1 Literature Search Strategy 16
2.2 Community Bystanders’ Beliefs 17
2.3 Bystander Intervention Programmes 18
2.4 Bystanders during Covid-19 18
3.0 Methodology 19
3.1 Research Question and Aims 19
3.2 Methods Overview 19
3.3 Material Development 20
Survey 20
Interviews 20
3.4 Inclusion Criteria 21
3.5 Ethics 21
3.6 Dissemination 21
3.7 Data Analysis 21
4.0 Results 22
4.1 Survey Respondents 22
4.2 Interview Respondents 22
4.3 Bystander Experiences 23
4.4 Bystander Actions, Motivations and Barriers 26
4.5 Sexist ‘Banter’ 31
4.6 Sexual and Domestic Violence ‘Banter’ 31
5.0 Discussion 32
5.1 Key Findings 32
5.2 Limitations 35
5.3 Future Actions 36
Policy Options 36
Practice Options 37
Research Options 37
6.0 Concluding Comments 38
References 39
Appendices 43
A: Participant Demographics 43
B: Data Tables for DVA Witnessed During Pandemic 44
C: Data Tables for Sexist Banter 61
D: Data Tables for Sexual And Domestic Violence Banter 63
Contents

TABLE 1: Frequency and percentage of participants who witnessed 25
the different DVA behaviours
TABLE 2: Frequency of participants from each age group who took action 26
TABLE 3: Percentage of people who took action when accounting 27
for gender of perpetrator and victim
TABLE 4: Signicantrelationshipsbetweenbystandertraits     29
and motivations/barriers to taking action
TABLE 5: Participant demographics 43
TABLE 6: Frequency of behaviours witnessed 44
TABLE 7: Chi-squaredndingsforbehaviourswitnessed     45
and participant demographics
TABLE 8: Chi-squared results for participant demographics and 47
taking action against DVA
TABLE 9: Chi-squared results for details about DVA and taking action 50
TABLE 10: Chi-squared results for actions, motivations 52
and barriers for each DVA behaviour witnessed
TABLE 11: Chi-squared results for participant demographics 55
and motivations for taking action
TABLE 12: Chi-squared results for participant demographics 58
and barriers to taking action
TABLE 13: Chi-squared results for participant demographics 61
and witnessing sexist banter
TABLE 14: Details of the sexist banter witnessed 62
TABLE 15: Chi-squared results for participant demographics 63
and witnessing sexual and domestic violence banter
TABLE 16: Details of the sexual and domestic violence banter witnessed 64
FIGURE 1: Behaviours witnessed during the pandemic 23
FIGURE 2: Means by which the bystander initially became aware of the DVA behaviour 27
List of Tables
List of Figures

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
5
Acknowledgements
Firstly, we would like to thank all research participants without whom this study would
not have been possible.
We would also like to thank all those who supported the study whilst obtaining
approvals. These include Public Health Wales colleagues from Quality Nursing and
Allied Health, Risk and Information Governance, and the Research and Evaluation
Division.
Advisory Group: The team would like to thank the advisory group members for
their support throughout this research project; Joanne Hopkins (Public Health Wales),
Natalie Blakeborough (Public Health Wales), Miriam Merkova (Welsh Women’s
Aid), Amy Jones (Welsh Government), Alexa Gainsbury (Public Health England and
University of Exeter), Karen Hughes (Public Health Wales), Gemma Woolfe (South
WalesPoliceandCrimeCommissionersOfce)andMikeTaggart(NorthWalesPolice).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
6
Acronyms
COVID-19 Coronavirus
CSEW Crime Survey for England and Wales
DVA Domestic violence and abuse
HCRW Health and Care Research Wales
HRA Health Research Authority
ONS OfceforNationalStatistics
PHW Public Health Wales
VAWDASV Violence against women, domestic abuse and sexual violence
WHO World Health Organisation

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
7
Executive Summary
Background
Domestic violence and abuse (DVA) is a major
public health, human rights and criminal
justice concern. The COVID-19 pandemic has
exacerbated conditions for DVA with experts and
academics sharing concerns over the safety of
victims and survivors and accessibility of support
(Sacco et al., 2020; Speed et al., 2020). During
this period, contacts to helplines for DVA, such as
the Live Fear Free helpline in Wales have surged
(ONS, 2020c).
The changes in daily routines experienced by the
nation as a result of the pandemic has resulted
in different groups of people becoming aware
of DVA. This is evidenced by the increase in
third parties reporting concerns to the police or
domestic abuse helplines (Ivandic et al., 2020;
ONS, 2020). This study, conducted by the Wales
Violence Prevention Unit (VPU) and University of
Exeter, explores the experiences and behaviours
of bystanders to DVA during the COVID-19
pandemic.
Methodology
This mixed methods pilot study sought to explore
the following research questions:
1. What are bystanders’ experiences of
witnessing DVA during the COVID-19
pandemic?
2. What are the motivations and barriers
for bystanders taking action to prevent
DVA during the COVID-19 pandemic?
3. What was the impact on the bystanders
and what support do they need?
The online survey was developed by the research
teamspecicallyforthisstudy,withtheaimof
adapting it for future use with a wider target
audience and/or outside of the pandemic. The
survey covered demographic information,
personality traits, DVA witnessed since the
pandemic began, actions taken, motivations
and barriers to taking action, and impact of
the experience on the bystander. A total of 186
completed survey responses were analysed for
this study. The data was analysed using IBM
SPSS Statistics, version 24.
The interviews were offered to all those who
participated in the online survey. The interviews
followed a similar structure to the survey, but
asked participants to offer more details of their
experience. Three interviews with bystanders
werecompletedandusedtosupportthendings
of the survey.
This study used an online survey and follow-up interviews
with survey participants, through online platforms. Data
collection took place over a three week period, between
15th February 2021 and 8th March 2021. Participants were
asked to share their experiences since the pandemic
lockdowns began in Wales (March 2020).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
8
Results
The results highlighted that:
• The circumstances of the pandemic allowed
bystanders to become aware of DVA with
coercive controlling behaviours causing most
concern among participants (see section
4.3).
• The majority of abuse noticed by bystanders
in the study was within current or ex
intimate partner relationships, with women
more likely to be victims and men more
likely to be the perpetrator of the abuse
(see section 4.4).
• Feeling connected to their community was
 asignicantpredictorofthebystander 
taking prosocial action in response to the
behaviour that had caused them concern.
• Most participants offered to support the
victim and felt that they had the correct
skills to be able to do so.
• Those who did not take action indicated
that this was down to a lack of skills and
not knowing what to do.
• The experience of witnessing or being
concerned about DVA had a negative
impact upon survey respondents, yet most
would not have liked to do anything more
when thinking back.
• Survey respondents who said that they
had witnessed sexist banter or jokes
since the pandemic began had shared their
disapproval with the person saying it
(see section 4.5).
• A quarter of survey respondents had noticed
an increase in domestic or sexual abuse
jokes since the pandemic began. These
bystanders indicated that they took action
against this behaviour because they
recognised that it was problematic
(see section 4.6).
Future Actions
Policy Options
• This study demonstrates that bystanders
have an important role in the primary
prevention of DVA. This could be recognised
in violence prevention policy. Encouraging
prosocial behaviour, when safe to do so, could
be a priority both generally and particularly
during a pandemic or other emergency
situations, when services are not as readily
accessible. Policymakers could consider
the use of public awareness campaigns and
training to promote knowledge about DVA,
and prosocial and informed bystander
behaviour. This could help mitigate any
negative impact that the experience may have
on the bystander themselves (see section 4.4).
• Lessons from this research suggest
that public-facing bystander campaigns
should be multi-faceted and should be
underpinned by awareness and knowledge
raising of what constitutes DVA for a
public-facing audience. As noted in section
4.4, most participants had witnessed DVA
behaviours within intimate relationships,
when the perpetrator was a man, and the
victim was a woman. Campaigns
should emphasise that abuse can happen in
a variety of relationships, regardless of gender
identity, sexuality, age or ethnicity.
• Policy makers should consider how they
can engage different target audiences in
knowledge and awareness raising,
and as prosocial bystanders. In particular,
as noted in section 4.1, most participants in
this study were women. Particular attention
should be paid to ensuring men engage
in bystander efforts. Awareness raising
campaigns should aim to increase a sense
of responsibility and motivation to act
and therefore be accompanied by the offer
of evidence-based bystander training to
enable and empower bystanders to move
(safely) through the theory of behaviour
change (see section 5.1).
• The data indicated that a sense of
 communitywasfoundtobeasignicant 
predictor of bystanders taking prosocial
action against DVA (see section 4.4).
Therefore, policies could aim to nurture,
sustain and further encourage this sense
of community as a contributory factor,
motivating bystanders to taking action
against DVA.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
9
Practice Options
• The research demonstrated that there is
a demand for bystander training programmes
to empower and upskill bystanders to take
prosocial action (see section 4.4). Bystander
training programmes must be evidence and
theory-based and must take people through
the process of change (see section 2.3).
• Social norms theory should be incorporated
into bystander training programmes, materials
and campaigns. For an example of a bystander
training programme, see the DVA bystander
intervention, Active Bystander Communities
(see section 2.3). As noted in section 4.4, the
bystander feeling that they possess the correct
skill set to take action is essential. A variety
of bystander responses should be incorporated
into campaigns, materials and bystander
training programmes. These could be rolled
out across communities as bystanders may be
essential in the primary prevention of DVA, both
in and out of a pandemic.
• When considering public awareness campaigns
targeted at bystanders, where appropriate,
organisations should make clear which services
they offer that might be of relevance to
bystanders. As discussed in section 5.1,
bystanders are more likely to share their
concerns if they know how to. Clear signposting
to relevant bystander services would
allow bystanders to have increased chance of
building knowledge about what is available.
• An increasing number of calls are being
recorded to domestic abuse helplines and the
police from concerned third parties (neighbours,
friends and family) (see section 1.4). Findings
from this study indicate that bystanders are
often negatively impacted by their experience
(see section 4.4). Frontline services, including
the police and specialist DVA services, could
consider developing guidance and training
 forcallhandlersandrst-responderstosupport
bystanders who make contact.
Research Options
• This pilot study has tested the methodology,
dissemination and topic area of bystanders
to DVA during the COVID-19 pandemic.
Dissemination and survey recruitment
should be amended in future research, to
optimise the recruitment of men, BAME groups,
and elderly people (see section 5.2). Recruitment
should also run for a longer period of time
to optimise uptake and on a larger scale with a
population level sample.
• The survey respondents indicated that
their experience of taking action had a negative
impact upon themselves, yet more than half
indicated that there was no further actions
they wished they had taken (see section 4.4).
Future research could explore how these
negative impacts could be mitigated with
bystander training programmes and/or other
adequate support resources.
• Those with a greater sense of community
 weresignicantlymorelikelytotakeaction
in response to their concerns (see section 4.4).
Future research should determine what “sense
of community” means to each participant, such
as locality, religion, sports group etc. and how
 thissenseofcommunityreectsonthetypes
of behaviour witnessed or the types of action
taken.
• Further research could also explore the inherent
 difcultyinaskingpeopletoparticipatein 
research about the DVA they may have
witnessed when they may not recognise what
behaviours constitute DVA (see section 5.1 for
 discussiononthis).Thiscouldbeachievedrstly
by increased public awareness of what DVA is,
and secondly, by alternative methods of
recruitment campaigning which could offer
 morein–depthdenitionsandexamplesof 
behaviours.
An increasing number of calls are being
recorded to domestic abuse helplines and
the police from concerned third parties
(neighbours, friends and family).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
10
Conclusion
This study sought to explore the experiences and behaviours of bystanders to
DVA during the COVID-19 pandemic through a mixed methods approach with the
general public, including survey and interviews. Whilst implemented on a small
scale, this study was the rst of its kind and provides new insights into bystander
experiences during a global pandemic.
Findings from this study suggest that the
circumstances of the pandemic have increased
people’s opportunity to be active bystanders to
DVA behaviours. Participants reported being more
aware of ‘concerning’ behaviours due to increased
time spent at home, coupled with less ‘distraction’
from the norms of regular social and work life, and
a heightened sense of community. Participants
also felt that the circumstances of the pandemic
had increased the ability of perpetrators to control
the victim, with coercive control being the most
commonly witnessed behaviour.
Having received DVA training, or feeling that
they possessed the correct skills to take action
was a strong predictor of prosocial bystander
responses. Bystanders indicated that the provision
of information and training for bystanders would
be helpful to mitigate barriers to taking action
and guide them in how to take prosocial action
against DVA.
Domestic violence and abuse (DVA) is a major
public health, criminal justice and human rights
issue. It is a signicant cause of ill-health and
inequality, and has adverse social, psychological
and economic impacts for individuals, families and
communities across the life course (WHO, 2021).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
11
1.0 Introduction
1.1 Background of Domestic Violence
and Abuse in Wales
Domestic violence and abuse (DVA) is a major
public health, criminal justice and human rights
issue.Itisasignicantcauseofill-healthand
inequality, and has adverse social, psychological
and economic impacts for individuals, families
and communities across the life course
(WHO, 2021). Living without fear of violence and
abuse is a fundamental requirement for health
and wellbeing. The National Institute for Health
and Care Excellence (NICE, 2014) states that:
"The cost, in both human and economic terms,
is so signicant that even marginally effective
interventions are cost effective".AHomeOfce
report estimating the economic and social
costs of DVA in England and Wales, placed
the annual cost at £66 billion, with 71% of that
being attributed to addressing the physical and
emotional harm experienced by victims
(Oliver et al., 2019).
DVA can have fatal outcomes. Every day,
137 women are killed worldwide by a family
member. It has been estimated that more than
half (50,000) of the 87,000 women who were
intentionally killed in 2017 were killed by family
members or intimate partners. More than a third
of these women (30,000) were killed by a current
orexintimatepartner(UnitedNationsOfce
on Drugs and Crime, 2019). Between March
2018 and 2019, Welsh police forces recorded
80,924 DVA related incidents (ONS, 2020a), yet
recorded police data only highlights a fraction of
the real picture, as incidents often go unreported.
It is estimated that a total of 2.3 million adults
aged 16-74 living in Wales and England have
experienced DVA in the past year (ONS, 2020a).
Anyone can experience DVA, regardless
of gender identity, age, sexuality, ethnicity,
occupation and income. However, understanding
DVA requires an appreciation that it is part of a
social pattern of male violence towards women
(Hester and Lilley, 2014), with data illustrating
that it is predominantly women and girls who
are victims and survivors of DVA perpetrated
by men and boys (ONS, 2020b). Women and
girlsaresignicantlymorelikelytoexperience
severe forms of abuse, including physical and
sexual violence, which result in injury or death
(Hester, 2013). Furthermore, they are more likely
to experience repeated physical, emotional or
psychological abuse. Between 2016 and 2018,
270 out of 366 domestic homicide victims, in the
UK, who were killed by a current or ex intimate
partner were female (ONS, 2019).
1.2 Denition
Domestic violence and abuse is dened as “any incident
or pattern of incidents of controlling, coercive, threatening
behaviour, violence or abuse between those aged 16
or over who are, or have been, intimate partners or
family members regardless of gender or sexuality. It can
encompass, but is not limited to, the following types
of abuse: psychological, physical, sexual, nancial or
emotional” (UK Government, 2013).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
12
1.3 The Policy Context in Wales
Domestic violence and abuse is a human rights
issue, recognised in national and international
treaties and conventions, a criminal justice issue,
and a public health issue. The prevention of
violence against women is a priority for:
• The United Nations (UN), through the
‘Convention on the Elimination of All Forms of
Discrimination against Women’
• The European Union (EU) through the
‘Istanbul Convention’
• The UK Government through the ‘Strategy to
End Violence Against Women and Girls
2016-2020’
• The UK Government through the ‘Strategy
to End Violence Against Women and Girls
2021-2024’
• The Welsh Government through the
‘Violence against Women, Domestic Abuse
and Sexual Violence (Wales) Act 2015’
TheWelshGovernmentpublisheditsrst
national strategy in 2005. ‘Tackling Domestic
Abuse’ (Welsh Government, 2005) adopted a
rights-based framework guaranteeing every
citizen the right to live free from violence and
abuse (McCarry et al., 2018). The ‘Right to be Safe
Strategy’ followed. This six-year plan focussed
on four key areas; prevention, awareness raising,
supporting victims, and improving the response
of criminal justice services as well as health (and
related) services (Welsh Government, 2010).
In 2012, the Welsh Government white paper
set a course for improved education, awareness
and more integrated services (Welsh Government,
2012).
In 2015, the Welsh Government passed the
‘Violence against Women, Domestic Abuse and
Sexual Violence (Wales) Act’ (hereafter, the Act);
therstpieceoflegislationintheUnitedKingdom
to explicitly address violence against women as
opposed to domestic violence generally. The key
purpose of the Act was to improve the public
sector response in relation to the prevention of
acts of gender-based violence, domestic abuse,
and sexual violence, the protection of victims and
supportforthoseaffected.TheActissignicant
for women because it sets out practical steps
which national and local government and public
sector bodies should implement to work together
to prevent violence against women. The Act also
aims to strengthen the support available to the
victims of violence against women, domestic abuse
and sexual violence (VAWDASV) by improving the
public sector response and consistency of service
provision by providing for a strategic focus with a
preventative approach (Price et al., 2020, Jurasz,
2018).
In 2015, the Welsh Government passed the
‘Violence against Women, Domestic Abuse and
Sexual Violence (Wales) Act’, the rst piece of
legislation in the United Kingdom to explicitly address
violence against women as opposed to domestic
violence generally.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
13
1.4 COVID-19 and Domestic Violence
and Abuse in Wales
Since the Coronavirus (COVID-19) pandemic
began, countries worldwide have implemented
various measures to limit its impact. For those
living in Wales, these measures included social
distancing, isolation and lockdown, and closure of
retail outlets, pubs and bars, leisure facilities, places
of worship and some public places. People were
told to work from home wherever possible and only
leave for essential shopping and daily exercise.
These restrictions have varied in intensity and
design since the initial Welsh lockdown in March
2020. Whilst they are intended to keep the country
safe, for victims of DVA, they may have the exact
opposite effect (Campbell, 2020).
For those experiencing DVA, the pandemic has
created a “perfect storm” as conditions for DVA
have been exacerbated (Welsh Women’s Aid,
2020). While people are encouraged to work from
home and in “lockdown”, victims have been forced
to stay with their abusers for extended periods of
time with limited ability to access support or leave
the situation. This has given abusers increased
proximity to the victim, the opportunity to control
phoneuse,internetuse,nances,preventaccess
to medical services (A&E, contraception etc.), limit
contact with other people and stop the victim
accessing support networks (Sacco et al., 2020;
Speed et al.,2020;KofmanandGarn,2020).It
is likely that abusers’ behaviours may have been
further exacerbated by the psychological strains
caused by the pandemic, including concerns about
nancialsecurity,employmentandchildcare
(Snowdon et al., 2020; Kaukinen, 2020).
Experts suggest that a “shadow pandemic” has
occurred alongside the COVID-19 pandemic, as
levels of DVA have been increasing “behind closed
doors” (UN Women, 2020). At the beginning of
the pandemic, the WHO issued guidance to policy
makers urging them to ensure members of the
public knew what services were still available
to them concerning DVA (Pearson et al., 2021).
Concerns logged with the Welsh Women’s Aid Live
Fear Free Helpline1 haveincreasedsignicantly.
This rise has been noted across all channels: calls
(41%), emails (66%), webchats (10%) and texts
(768%) (ONS, 2020c). The Live Fear Free helpline
data also indicates an increase in third parties
(neighbours, friends, family, colleagues) contacting
the helpline for support and advice on what they
can do about someone they are concerned about
(ONS, 2020c) and there has been an increase in
the length, severity and complexity of calls (Wales
Violence Prevention Unit, 2020). These trends are
supported by Ivandic et al. (2020) who found an
increase in domestic abuse calls to the police from
third parties, especially in high density areas. This
suggests that the “stay at home” guidance has
not only increased but exposed new and different
groups of people to witnessing or having concerns
about DVA.
1 The Live Fear Free Helpline is a free helpline for anyone living in Wales experiencing or with concerns about VAWDASV
Concerns logged with the Welsh Women’s Aid Live Fear Free Helpline
have increased signicantly. This rise has been noted across all channels:
41%
Calls
10%
Webchats
768%
Texts
66%
Emails

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
14
1.5 Bystanders to Domestic Violence
and Abuse
Bystanders are “witnesses to negative behaviour (an
emergency, a crime, a rule violating behaviour) who,
by their presence, have the opportunity to step in to
provide help, contribute to the negative behaviour or
encourage it in some way, or stand by and do nothing
but observe” (Banyard, 2015, pp. 8). Everyone is a
bystander, all the time.
We witness events unfolding around us constantly.
Some of these events may be recognised as
problematic, and we might decide to do or say
something, becoming an active bystander, or to do
nothing and remain a passive bystander. There are
manyfactorsthatwillinuencewhywedecideto
intervene or not (Fenton et al., 2016).
Noticing the event
1. 2.
3. 4.
Possessing the right skills
to take action (Latane and
Darley 1979, Berkowitz, 2009)
Recognising the
situation as problematic
(in this case, recognising
the situation as DVA)
Feeling responsible
to take action
These steps are at the core of most bystander
training programmes (e.g. Gainsbury et al., 2020);
if at any point a bystander does not identify with
a step, for example, they do not recognise the
situation as DVA (step 2), they may not take any
action in response to the behaviour they have
noticed.
Socialnormstheoryalsoplaysasignicantrole
in bystander responses. When applied to DVA,
this means that social or communal norms can
inuencethewaythatperpetratorsandbystanders
behave. “It is not necessary for the majority [of
the group] to believe it, but only for the majority to
believe that the majority believes it” (Berkowitz,
2003, pp. 261). Put simply, a bystander is less likely
to take action, if they become concerned about
DVA, when no one else seems concerned about the
behaviour. Other people may also be concerned,
butnotfeelcondenttosayanythingasthegroup
norm implies that it is not a problem; this is often
referred to as “pluralistic ignorance”. Shifting the
social norms, and empowering bystanders to feel
condentenoughtospeakup,isthereforeessential
to tackling DVA (Fenton et al., 2016).
The theory of change that inuences active bystander responses is

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
15
When DVA is not reported, it can have severe
consequences for the victim; the most serious
of which is death (Bouillon-Minois, 2020). Yet
researchers have indicated that bystanders are
more likely to report instances of DVA when they
have proof, because they are worried that they
will not be believed (Rowe, 2018). It is therefore
imperative that bystanders are provided with
enoughinformationtofeelcondentthattheir
concerns, even without proof, will be taken
seriously and that the methods for sharing their
worries are easily accessible (Borum, 2013).
Bystanders are essential to tackling DVA at a
community and societal level (Gainsbury et al
2020). Their support can also be vital for victims
living with DVA.
1.6 Study Rationale and Aims
DVA can often be a crime that occurs behind closed
doors and out of sight. The restrictive measures
put in place to contain the spread of COVID-19
have meant that for some victims, DVA has
been exacerbated and means to access support
networks has been restricted. In turn, other people
in physical (and online) proximity to the victims
may have become inadvertent and unexpected
bystanders to DVA and/or the warning signs of
abuse, with potentially new opportunities to
take action.
The National Strategy for VAWDASV (Welsh
Government, 2016) states that a key
commitment for Welsh Government is primary
prevention. This encompasses plans to increase
awareness of VAWDASV within the Welsh
population, and challenge the attitudes which
legitimise it. Additional key commitments
outlined in the strategy are building institutional,
organisational and community capacity to identify
and respond appropriately to VAWDASV, and
to adequately fund early intervention support
services. Bystanders to DVA may be an important
source of community resilience and support
for victims of DVA, perhaps more so during the
pandemic than ever.
This study aims to understand the experiences and behaviours of bystanders
during the COVID-19 pandemic. The study has the following objectives:
Provide insights into the
groups of people who have
witnessed DVA and its
warning signs
Identify the types of abuse
bystanders are witnessing
Explore the actions taken by bystanders in response to the DVA
concerns, and the motivations or barriers to taking that action
The ndings from this study can be used to inform DVA prevention policy,
the development of bystander training programmes and improve knowledge
about bystander responses to DVA during the COVID-19 pandemic.
Understanding these experiences and behaviours is crucial for creating
societal resilience, both during the COVID-19 pandemic and in the future;
and for the prevention of DVA as a critical public health strategy.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
16
2.1 Literature Search Strategy
A scoping review was used to examine the
relevant literature. Scoping reviews are useful to
• Identify what evidence is available,
• Examine the methodologies used,
• Identify key factors related to the research subject
• Identify any knowledge gaps
(Munn et al., 2018).
The review aimed to identify literature which
supports the research objectives (as above).
Thearticlesidentiedwereinitiallyusedto
inform the development of the survey, and
latertocritiquethendings.
Eleven search engines were utilised for the
search: Wiley, Taylor and Francis, Springer, Sage
Journals, ProQuest, Elsevier, Cochrane, CINAHL,
APA, Research Gate and Google Scholar. This
was accompanied by a hand search through
relevant journals; “Violence Against Women”,
“Journal of Gender Based Violence”, “Psychology
of Violence”, “Journal of Interpersonal Violence”
and “BMC Public Health”. The key search terms
used were “bystander”, “bystander experiences”,
“COVID”, “coronavirus”, “pandemic”, “domestic
abuse and violence”, “domestic”, “abuse” and
“violence”.
Search limiters included English language only,
peer reviewed and full text being available.
Published systematic reviews were also explored
with relevant references being followed up.
The scoping review was conducted between
November 2020 and April 2021.
2.0 Literature Review

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
17
2.2 Community Bystanders’ Beliefs
Severalstudieswereidentiedthatexplorethe
beliefs, behaviour and perceptions of bystanders to
a variety of different DVA stimulus.
Banyard et al. (2020) sent postal questionnaires
to residents in New England, US to explore any
links between sense of community and prosocial
bystander responses to domestic and sexual
violence. The 1,623 respondents were divided
into three categories: non responders, occasional
responders, and frequent responders. All of those
within the frequent responders group had an
increased sense of community compared to the
other two groups. Whilst not conducted within a
community setting, McMahon et al. (2015) found
corresponding results within a university sample.
Muralidharan and Kim (2019) sought to determine
themostefcientmeansofmotivatingbystanders
to take action against DVA. They used two groups
of participants. One group was taught facts about
DVA and the other was presented with a survivor’s
lived experience narrative. The lived experience
narrativewassignicantlymoreeffectiveat
motivating bystanders to theoretically take
prosocial action against DVA. This stark difference
between the groups was attributed to the empathy
that bystanders were able to feel after hearing
the victim’s personal experience2. This suggests
that empathy plays an important role in prosocial
bystander behaviours. Empathy is an antecedent
toefcacy,wherebytheonelogicallyprecedes
theother.Thelinkbetweenincreasedefcacyand
prosocial bystander intervention is well established
(Fenton and Mott, 2018).
Similar to the above study, Green (2020) showed
villagers in rural Uganda videos of DVA victims
sharing their stories to try and reduce violence
against women and encourage bystander
intervention. Several months after showing the
videos, the research team revisited the villages to
assess the impact of the videos. Rates of violence
against women were unaffected by the videos,
yet more people were reporting incidents they had
seen or personally experienced. This suggests that
empathy is more effective at motivating prosocial
bystander responses and encouraging victims to
come forward than it is at deterring abusers.
Taylor et al. (2016) explored the experiences of
victims of intimate partner violence (IPV) in rural
areas of the Southern United States when a
bystander was present to witness the physical
abuse. A stark number of participants indicated
that the presence of the bystander resulted in
higher rates of injury and poorer victim mental
health. Although solely focused on physical IPV,
this study highlighted that not all abuse occurs in
private, and when an (untrained) bystander had
been present, it resulted in worse outcomes for the
victim. Bystander interventions must ensure that
safety for themselves (as a bystander) and the
victim is a priority when trying to help in a situation
related to physical violence. This study has
highlighted why appropriate training is essential, to
protect the safety of all involved.
Despite not being conducted within the general
community, several studies have shed light on
motivations and barriers to students taking
action when they witness domestic or sexual
abuse. Whilst not directly relevant to this study,
they still offer insight into bystander behaviours.
For example, Weitzman et al., (2020) found that
the relationship between the bystander and the
victimwasasignicantpredictoroftakingaction
or not, with participants being most likely to take
action if the victim was a friend or family member.
Flemming and Wiersma-Mosley (2015) found that
the severity of the abuse being witnessed was
highlyinuentialforthebystandertakingaction.
The students were much more likely to report
incidents where they felt there was an immediate
threat to life. Christensen and Harris (2019) found
that bystanders within the student population were
much more likely to take prosocial action if they had
personal experience of being a victim themselves.
2 Listening to a victim’s story can sometimes have the adverse effect on male attitudes towards violence against
women; for example, Berg et al. (1999) found that male undergraduates reported increased likelihood of engaging in
rape-supportive behaviours after hearing a female rape victim’s narrative.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
18
2.3 Bystander Intervention Programmes
Thereisasignicantglobalbodyofliteratureon
bystander programmes to prevent domestic and
sexual violence and abuse. In a recent systematic
review, conducted by Addis and Snowdon (2021),
bystander programmes were the focus of seven
systematic reviews (Kovalenko et al., 2020; Kettrey
and Marx, 2019a; Kettrey and Marx, 2019b;
Mujal et al., 2019; Jouriles et al., 2018; Storer et
al., 2016; Fenton et al., 2016). All reviews found
that the majority of bystander literature focuses
on adolescents or young people in educational
settings, mainly college and university settings with
a focus on sexual and intimate partner violence
prevention. Overall, the review concluded that
bystander programmes have a strong aptitude
for changing attitudes and beliefs that promote
VAWDASV.
Additionally, seven primary studies evaluated
bystander interventions including the Red Flag
Campaign (Borsky et al., 2018, Carlyle et al.,
2020), Green Dot (Coker et al., 2019), Bringing
in the Bystander (Edwards et al., 2019) and The
Intervention Initiative (Fenton and Mott, 2018).
While studies were predominantly undertaken
in university settings, one study indicates that
the bystander approach (Active Bystander
Communities) can be transferred from student
populations to general communities in the UK
(Gainsbury et al., 2020).
Gainsbury et al. (2020) evaluated the feasibility
and potential for effectiveness of a DVA bystander
intervention within UK communities. Active
Bystander Communities (ABC) require participants
to attend a training programme facilitated by
expertsoverthreesessions.Bystanderefcacy,
behavioural intent, bystander behaviours and
myth acceptance were assessed at baseline
using validated scales, post training and after
fourmonths.Signicantchangeswereobserved
acrossbystanderefcacy,behaviouralintentand
myth acceptance. At four months post intervention,
these changes had been maintained, and in
some cases had increased, with the exception of
mythacceptance.Thesendingsarepromising
and indicate that ABC can help change attitudes
towards DVA in general communities within the UK.
Additionalstudiesidentiedprogrammeswith
elements of bystander programming built into
programme design. For example, Quigg and
Bigland (2020) conducted an evaluation of The
Good Night Out Campaign (GNOC). The GNOC is
a UK programme that was developed for licenced
premises which aims to support those who
work in nightlife settings to better understand,
respond to, and prevent sexual violence. GNOC
facilitators worked with 11 nightlife venues,
providing guidance on preventing and responding
to sexual violence and bystander training for over
150 nightlife workers. The participating venues
were provided with materials to display to raise
awareness of the GNOC and encourage patrons
to report incidents. The trainees were surveyed
andndingssuggestthattheGNOCtraining
programme is associated with improvements in
knowledge and attitudes towards sexual violence.
Quigg and Bigland (2020) also noted greater
readinessandcondencetointerveneinsexual
violence amongst nightlife workers.
2.4 Bystanders during COVID-19
Onlyonestudywasidentiedwhichmade
reference to bystander behaviours during the
pandemic. Campbell (2020) noted that only 8%
of calls to law enforcement agencies to report
DVA during the pandemic came from bystanders,
compared to 80% of animal control calls which
were made by bystanders. This paper further
highlights how DVA is also present in a large
number of the homes where animal cruelty is
reported; a link that is now well established
(Riggs et al., 2021). Animal control units
should also be trained in DVA and be made
aware of the routes to report concerns they
notice whilst investigating cases of animal abuse.
There is a clear gap in the evidence base for
bystander behaviours during the COVID-19
pandemic, an opening this study sought to
address.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
19
3.0 Methodology
3.1 Research Question and Aims
This research seeks to explore the behaviours
and experiences of bystanders to DVA during the
COVID-19 pandemic in Wales. In doing so, this
study will address an understudied area in the
bystander and violence prevention literature. The
following research questions were posed:
• What are bystanders’ experiences of DVA
during the COVID-19 pandemic?
• What are the motivations and barriers
for bystanders to DVA during the COVID-19
pandemic?
• What was the impact on bystanders and
what support do they need?
The aims of this study were to:
• Improve knowledge of bystander
opportunities and behaviours during the
COVID-19 pandemic.
• Inform policy and prevention strategies.
• Add to the evidence base for bystander
programmes and how the primary
prevention of DVA can be utilised during
future pandemics.
3.2 Methods Overview
This mixed methods study used an online quantitative survey
delivered through the online platform Qualtrics, combined
with qualitative interviews, conducted over Zoom or
Microsoft Teams. Survey and interview recruitment ran for a
three week period, from 15th February 2021 to 8th March 2021.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
20
3.3 Material Development
Survey
To the best of the research team’s knowledge,
no survey or questionnaire currently exists which
explores the experiences of bystanders to DVA
during the COVID-19 pandemic. Therefore, the
team developed their own survey with the help of
the advisory group (acknowledged above,
pp 5). To develop the survey, an extensive search
of the literature and campus climate surveys was
conducted, alongside a review of the current data
trends and criminal law. Participatory workshops
were held with the research team and members
oftheadvisorygrouptoshapeandrenethenal
survey. An academic paper outlining the process of
developing this innovative survey instrument is due
to be published shortly.
Details on the survey dissemination can be found
below (Section 3.6, pp 22).
Interviews
Semi-structured interviews were conducted with
participants via Microsoft Teams or Zoom. These
interviews followed a similar structure to the online
survey but also asked participants to expand on
the impact of being a bystander to DVA had on
them personally, and those involved. The interviews
allowed the participants to share the holistic details
of what they had witnessed or become concerned
about since the pandemic began. They were able to
paint a fuller picture of the events which the survey
was unable to capture. A copy of the interview
questions is available upon request.
On average, each interview lasted half an hour.
Findings from these interviews have been entwined
with the survey data, to further illustrate key
ndings.
The nal version of the survey contained the following sections:
Demographic
information
Knowledge and
attitude of DVA
DVA witnessed
during the pandemic
Sexist and
misogynistic
jokes or banter
witnessed during
the pandemic
Sexual violence
and domestic
abuse, jokes
or banter
witnessed during
the pandemic
Training needs
Findings from the online survey can be found below (Section 4.0, pp 22).
At the end of the survey, respondents were asked if they would like to take part
in an interview to explore their experiences further. If so, they were invited to
email the research team and request an information sheet which contained
further information on the interviews. If they wanted to take part in the
interview after reading through the information sheet, this was also arranged
through email with the research team.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
21
3.4 Inclusion Criteria
This study aimed to recruit anyone aged 18 or
over who may have seen or become aware of
DVA since the beginning of lockdown (March 2020).
This study was solely conducted in Wales with
participants living or working in the country.
3.5 Ethics
After consulting with Public Health Wales’
Safeguarding Lead, this study received
NHS ethics approvals from HRA and HCRW
on 20th January 2021 (ref. 20/HCRW/0061).
Public Health Wales’ Data Protection and Public
Health Wales Research and Development
approved the study on 4th February 2021.
3.6 Dissemination
A communication plan was developed to advertise
and optimise the reach of the survey across public
and professional networks, using both remunerated
and organic communications through social media
and email. Key organisations and stakeholders
were engaged to aid in the dissemination of the
survey through public channels and professional
networks.
When designing the advertisements for recruitment
into this study, the word “bystander” was not
used, as the team were not sure members of
the public would know what this word means.
Instead, the advertisements’ wording (from the
participant information sheet) was “a study into
what you might have witnessed, noticed or been
concerned about relating to domestic abuse or its
warning signs since the beginning of the COVID-19
pandemic”.
The (paid-for) communications were designed
for social media and included the design of an
advert in the form of a GIF (translated to Welsh
and English) which was used to advertise the
survey. Adverts were targeted at adults living
in Wales, certain occupation types who were
actively working within the community during
the pandemic; such as delivery/postal workers,
public transport workers, police, hospital staff,
hairdressers, and those who had visited DVA
websites. After two weeks, the data was reviewed
andanyidentiedgapsindemographicswere
targeted for a further week.
Communication also included an advertorial in
Wales’ largest online news site, Wales Online3. As
well as the advertorial, this included two Facebook
posts and two Twitter posts from the main Wales
Online newsfeed. The organic communications
included media and stakeholder engagement.
Coverage of the survey was also included on BBC
Wales Online4, BBC News webpage, BBC Radio
Wales’ morning radio show, and across the three
regional news cycles on the BBC.
3.7 Data Analysis
The quantitative survey data was initially cleaned,
with any partial responses being removed, and
codes were applied. It was then analysed using
IBM SPSS Statistics Version 24 and descriptive
statistical analysis and chi-squared were run.
A copy of the analyses outputs can be found in
Appendix B.
The interviews were audio recorded, transcribed
and anonymised. They were then analysed using
Atlas ti. 8 and thematic analysis was completed.
An academic paper on the results of the study
is due to be published in due course.
3 Take this survey and help tackle the rising cases of domestic abuse in Wales - Wales Online
https://www.walesonline.co.uk/special-features/take-survey-help-tackle-rising-19725171
4 https://www.bbc.co.uk/news/uk-wales-56074048

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
22
4.0 Results
This section combines the ndings from the quantitative survey and the
qualitative interviews. Where possible, the qualitative data is used to
support the quantitative data analysis and offer additional insight.
4.1 Survey Respondents
A total number of 395 survey responses were received over a
three-week period. Of these, 186 records were fully completed and
used for this report.
95.2% of respondents were White British/Irish which is consistent
with the demographics for Wales (Welsh Government, 2020). The
majority (85%) of the respondents were women (N=158 out of
186) and aged between 18 and 44 years old (66.3%).
A full breakdown of the demographic details of the participants can
be found in Appendix A.
The majority of participants (64%) indicated that
they felt their knowledge of DVA was very to
extremely knowledgeable. Participants did not
feel as knowledgeable about the laws relating
to DVA, with most selecting the moderately
knowledgeable option for this question on the
survey (52.7%). There was a nearly equal divide
between those who had attended DVA training
within the last 5 years, and those who had not.
There was a similar divide between participants
whose job roles required them to have awareness
of DVA and those who did not.
The survey captured data from people in a range
of occupations, including health care, social
care, industry, key public services, government,
education, and also included those who are
retired, unemployed and stay at home parents.
The data also indicated that three quarters
(75.4%) of the respondents had remained
primarily at home during the day since the
pandemic restrictions began; whether that
be working from home, retired, unemployed,
furloughed or stay-at-home parent.
Asignicantnumberofparticipantsindicated
that they were prosocial individuals, and wanted
to help other members of their community (80%).
Similarly, they felt that they could take action
against DVA, and did not agree with rape myth
statements.
4.2 Interview Respondents
Only six survey respondents emailed the team
to ask for more information about the interviews
and all six agreed to take part. Of these six,
three of the interviews were discounted from the
study as not meeting inclusion criteria- two were
survivors and one included experiences from
working in a professional capacity only.
The remaining three participants were women
who had good knowledge of DVA through their
own professional roles, and were able to discuss
concerns about DVA that they had outside of
their roles. Two had concerns about a friend,
while the other was concerned about their
elderly parents with dementia being abused by a
caregiver.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
23
4.3 Bystander Experiences
Coercive control behaviours and warning signs
of DVA were the most prevalent behaviours
witnessed by respondents (Figure 1). Coercive
control behaviours included someone being
monitored and/or controlled about where they are
or, who they are with; someone being monitored
using spyware or tracking devices; someone
having their phone, social media or internet use
controlled or checked. Warning signs of DVA
included someone looking fearful and/or walking on
eggshells around their partner and/or members of
their family; or someone who behaves as if they are
very worried and fearful all of the time.
This theme was also evident within the interview
data, as the interview participants had not
witnessed their friends or family being directly
abused, but became aware of warning signs which
indicated that something was not quite right.
Furthermore, interview participants felt that the
circumstances of the pandemic were being used by
perpetrators to control the victims.
“There was a denite change in my friend’s ability to talk freely... She was
starting to mention and drop things into conversations, so we were feeling
that maybe she wanted to talk about it …but as soon as you try to offer
support around that, it quickly shut down, very defensive” [Interview 3]
Figure 1: Behaviours witnessed during the pandemic
DVA behaviours witnessed by bystanders during the pandemic
Warning signs
24%
Coercive
30%
LGBTQI+
4%
Vulnerability
7%
Threats
14%
Physical
15%
Sexual
6%
"

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
24
Among the survey responses, there were
substantial gender differences noted within the
behaviours that had been witnessed and/or caused
concern. As Table 1 shows (pp 25), on the whole,
the majority of DVA behaviours were observed
by women, with the exception of the abuse of a
LGBTQI+ person, which was witnessed by 14.3%
of men participants, compared to 12.3% of women
participants, and threats of abuse, which was
witnessed by 57.1% of men participants, compared
to 41.1% of women participants.
For most behaviours, they were observed more by
those aged between 18 and 34 (as demonstrated
in Table 1), with the exception of LGBTQI+5 abuse,
the abuse of a vulnerable person and threats of
abuse, which was noted mostly by those over 45
years of age (52%, 42.5% and 39.4% respectively).
Exploration of a link between occupation and
behaviours witnessed showed a higher proportion
of those working in health and social care or
industry and other tertiary jobs6 reported becoming
concerned across all behaviours (see Table 1).
100% of participants who worked in education
(N=31) reported concerns about coercive control
since the pandemic began.
Although three quarters of respondents to the
survey were primarily at home, the data suggests
that there was no notable difference between those
who were primarily at home during the pandemic
(working from home, furloughed, unemployed) and
those who were continuing to go to work as normal
in witnessing or becoming concerned about DVA
behaviours (see appendices, Table 7). Yet notably
more participants (44.8%) indicated that they felt
the pandemic had allowed them to become aware
of the concerning behaviours (for example, working
athomewhentheywouldusuallybeintheofce).
This was also demonstrated in the interviews, as
participants felt that the change in daily routines
had allowed them to notice the warning signs.
5 Abuse aimed at lesbian, gay, bisexual, transgender, queer and intersex, + others.
6 Industry jobs include manufacturing or construction. Tertiary jobs included hairdressers, barbers, beauty therapists,
photographers, musicians, artists, transport, retail, hospitality and voluntary workers.
“She was quite fearful of the pandemic, he was using
that fear to keep her in the house more, to control her
more …she wouldn’t even go out into the garden in the
end because he was telling her it was wasn’t safe to do
that. So, it got quite extreme” [Interview 2]
"
"
“I think it would have been more easily hidden
or we might have been distracted from it and
we might not have been as proactive or as
aware and worried about it if we weren’t in
a pandemic” [Interview 3]

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
25
Table 1: Frequency and percentage of participants who witnessed the different DVA behaviours
Behaviours Witnessed
Warning
Signs
Coercive
Control LGBTQI+ Vulnerability Threats Physical Sexual
Man 15
(53.6%)
24
(85.7%)
4
(14.3%)
5
(17.9%)
16
(57.1%)
10
(35.7%)
2
(7.1%)
Woman 117
(74.1%)
144
(91.1%)
21
(13.3%)
35
(22.2%)
65
(41.1%)
73
(46.2%)
33
(20.9%)
18-34 52
(39.4%)
66
(39.3%)
7
(28%)
11
(27.5%)
24
(29.6%)
29
(34.9%)
13
(37.2%)
35-44 39
(29.5%)
49
(29.2%)
5
(20%)
12
(30%)
25
(30.9%)
25
(30.2%)
11
(31.4%)
45+ 41
(31.1%)
53
(31.5%)
13
(52%)
17
(42.5%)
32
(39.5%)
29
(34.9%)
11
(31.4%)
Usually at
home
19
(63.3%)
27
(90%)
3
(10%)
4
(13.3%)
15
(50%)
13
(43.3%)
4
(13.3%)
Health
and social
care
19
(63.3%)
27
(90%)
3
(10%)
4
(13.3%)
15
(50%)
13
(43.3%)
4
(13.3%)
Government
or public
services
26
(66.7%)
27
(90%)
3
(7.7%)
5
(12.8%)
17
(43.6%)
19
(48.7%)
7
(17.9%)
Education 25
(80.6%)
31
(100%)
1
(3.2%)
7
(22.6%)
11
(35.5%)
15
(48.4%)
5
(16.1%)
Industry or
tertiary
32
(71.1%)
37
(82.2%)
7
(15.6%)
16
(35.6%)
19
(42.2%)
16
(35.6%)
9
(20%)
GenderAge
Occupation

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
26
4.4 Bystander Actions, Motivations
and Barriers
Actions
Bystanders were asked if they had taken action in
response to the behaviour they had witnessed.
Results indicate that 85.7% (24 of 28) of men and
88.6% (140 of 158) of women took some form
of action. A chi–square test was run on these two
variables (gender and taking action) and no
signicantrelationshipwasfound(X2, p=.662).
The majority of participants in each age group took
action against the DVA they had witnessed
(see Table 2). The chi-squared analysis indicated
thattherewasasignicantrelationshipbetween
the age of the participant and whether or not they
tookaction,X2(2, N=186) = 6.296, p=.043.
Table 2: Frequency of participants from each age group who took action
When comparing bystanders taking action
(yes or no) to the respondents’ trait data, the
chi-square test showed that changing levels of
DVA awareness (increased or stayed the same)
duringthepandemicwassignicantlyassociated
withtakingactionagainstDVA,X2(1, N=186)
=4.330, p=.037. Analysis further indicated a
signicantrelationshipbetweentheparticipant
having attended a DVA training course and
takingactionagainstDVA,X2(1, N=186) = 6.311,
p=.012. Of the participants who had completed
a DVA training course, 94.4% also took action in
response to witnessing DVA, compared to 82.5% of
participants who had not received training, but still
took action.
All of those who witnessed the abuse of a
vulnerable person, threats of abuse or sexual
abuse reported that they had taken some form of
action in response. 86% of those who witnessed
coercive control and 74% of those who witnessed
warning signs of DVA took action. The majority
of participants took action after seeing the
behaviourmorethanvetimes(53.7%).Noneof
these relationships were shown to be statistically
signicant(seeappendices,Table9).
In the main, respondents showed that they
had become aware of the behaviour either by
witnessing it in person (N=75) or because the victim
told them directly (N=67). Six participants became
aware of the behaviour online, while 23 were told
by someone else. The remaining 15 did not wish
to answer (see Figure 2). The means by which the
behaviourwaswitnessedwassignicantlyrelated
towhetherthebystandertookaction,X2(4, N=186)
= 15.400, p=.004. This means that participants
weresignicantlymorelikelytotakeactionifthey
witnessed the problematic behaviours in person or
they were told about it directly by the victim, than
when they witnessed the abuse online or when
they were told by someone else.
Took action (yes)
Age
18-34 56 (81.2%)
35-44 48 (88.9%)
45+ 60 (95.2%)
Most participants who took action against the DVA behaviours
they had witnessed indicated that they wanted to help members of
their community (89.3%) and felt more connected to their community
since the pandemic began (89.3%). This suggests that those who took
action are prosocial people.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
27
Figure 2: Means by which the bystander initially became aware of the DVA behaviour
Most participants indicated that they had seen DVA behaviours in intimate
partner relationships (74.7%).Therewasnosignicantlinkbetweentypeof
relationship (intimate partner or family members) and the bystander taking action
(see appendices, Table 9).
The participants identied most of the perpetrators to be men (131 out of 186),
and most victims to be women (137 out of 186). Signicantassociationswerefound
betweengenderoftheperpetratorandtakingaction,X2(2, N=186) = 17.963, p<.001, and
thegenderofthevictimandthebystanderstakingaction,X2(2, N=186) =11.434, p=.003.
Yet, as Table 3 shows, there is only a small difference in the percentages of people
who took action when accounting for the gender of the perpetrator and victim.
Table 3: Percentage of people who took action when accounting for gender of perpetrator and victim
Took action (yes)
Gender of perpetrator
Man 131 (90.3%)
Woman 31 (88.6%)
Gender of victim
Man 25 (89.3%)
Woman 137 (89.5%)
Means by which the bystander initially became aware of the DVA behaviours
Did not wish
to answer
Told by
someone else
Online
Told by the
victim
Physically, in
person
010 20 30 40 50 60 70 80

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
28
Forthemostpart,therewerenosignicant
relationships between the type of behaviour
witnessed and the type of action taken by the
bystander, with two exceptions. These exceptions
tothetypeofactiontakenwereofciallyreporting
the incident to the police, DVA charity or social
services, and the bystander offering to support
thevictim.Thereweresignicantrelationships
between the behaviour witnessed and the
bystanderofciallyreportingtheincident7,
X2 (5, N=186) = 22.448, p<.001, and the
bystander offering to support the victim,
X2 (5, N=186) = 13.336, p=.020. Bystanders are
morelikelytoofciallyreporttheincidentifthey
witness coercive control (25.4%), threats (76.9%)
or physical violence (36%) rather than other forms
of DVA. Offering to support the victim was the most
common form of bystander behaviour reported
by participants, with 104 indicating that they had
done this (see appendices, Table 10).
The data shows that 96.2% of the bystanders who
took action knew that someone else was aware
of the DVA behaviours, compared to 87.2% who
did not know that someone else was aware of the
behaviour, but still took action. This offered another
signicantconnectionbetweenbystanderaction
and someone else seeing the behaviour,
X2(2, N=186) = 22.167, p<.001.
One interviewee explained that she had concerns
about her friend’s behaviour but was not sure. To
help her decide whether to say something, she
sourced allies. With other friends, she explained
her concerns, found that her other friends had also
noticed the warning signs and, together they came
up with a plan.
Motivations and Barriers
Two thirds (65.6%) of survey participants shared
the motivations they experienced when witnessing
coercive control. The primary motivation for taking
action was the bystander feeling responsible
(N=122), closely followed by recognising that the
behaviour was problematic and wrong (N=118).
Eight participants indicated that they had taken
action for “personal reasons”. Personal reasons
included being able to empathise with the victim
from personal experience and being personally
offended.
By contrast, the main barriers to taking action were
not recognising the situation as an issue (6 out of
24) and lacking the skills to intervene (4 out of 24).
Fourteen participants who did not take action gave
no reason as to why.
Dataanalysisrevealedthatthereweresignicant
relationships between many of the bystander traits
and the individual motivations and barriers for
taking action. For example, survey respondents
who said that they wanted to help members of
theircommunityweresignicantlymorelikelyto
also say that they helped the victim because they
recognised the situation as problematic or they
knew what to do to help. Table 1 illustrates the
otherstatisticallysignicantassociations
(see appendices, Table 11 for full breakdown).
7 To the police, DVA charity, social services etc.
“That informaiton gathering to see what other people [friends]
had noticed. It was more a case of nding out what their
experiences of it were so that I knew there was more to it than
just me thinking the worse...we tried to plan out the best way to
manage it [their concerns]” [Interview 3]
"
The data shows that 96.2% of
the bystanders who took action
knew that someone else was
aware of the DVA behaviours

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
29
Bystander Trait Motivation
I want to help members of my
community (prosocial person)
Recognising the behaviour as
problematic (p=.008)
Possessing correct skills
(p=.027)
Bystander knowledge of
DVA laws Possessing correct skills (p<.01)
Done course or training on DVA
in past 5 years Possessing correct skills (p<.01)
DVA awareness part of
occupational role
Possessing correct skills
(p=.001)
Feel more connected to
community since pandemic began
Possessing correct skills
(p=.004)
Bystander Trait Barriers
DVA knowledge
Pluralistic ignorance (other
people did not seem concerned
or the bystander was not sure
if other people would support
them) (p=.032)
Feeling more connected to the
community since the pandemic
began
Fear of retaliation (p=.027)
Table 4: Signicant relationships between bystander traits and motivations/barriers to taking action

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
30
All of the participants who indicated that they took
action (N=163) also indicated that they felt they
possessed the correct skills to know what to say
or do. Possessing the right skills is important when
being a prosocial bystander, as the interview
participants explained,
One interviewee explained that she had not taken any
action in response to the abuse she had witnessed
because she did not know how to report the person
without involving the police. Furthermore, she explains
that some perpetrators have psychological needs of
their own which may be recognised by the bystander
and prevent them from taking further actions.
Almost all participants explained that their experience
of witnessing or becoming concerned about DVA
since the pandemic began had negatively impacted
upontheirphysical,psychological,nancialorsocial
well-being. Only 8.1% of respondents felt that their
experience had a positive impact on them. When asked
if they would have liked to have done more in response
to their concerns, 50% of participants said “no” and a
further 10% did not answer the question. For the most
part, the remaining 40% indicated that they would
have liked to have supported the victim or reported
theincidenttoofcialbodies.75%ofrespondents
indicated that they thought having training on how to
help in situations related to DVA would be useful. This
also came across during the interviews where one
participant explained:
“It has played on my mind a lot, second guessing myself, did I say
the right thing? Did I push enough? Should I have pushed more?
And said we are really, really worried, maybe I pushed too much
by sending the message in the rst place. It has been a lot of
questioning myself” [Interview 3]
“This woman who has huge needs
of her own, so what I wanted to
ag up is before you black and
white say these are perpetrators
and these are victims, she is kind of
a victim. So I don’t want to throw
her into prison… unless I report it to
the police, which I think would be
cruel, what can I do?” [Interview 1]
“If you say the wrong thing to them,
it can have the adverse effect to
what you’re trying to do so you have
got to bite your tongue and be so
careful with what you say and do”
[Interview 2]
"
"
"

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
31
4.6 Sexual and Domestic Violence ‘Banter’
Participants were asked if they had noticed or become concerned about verbal
or written indicators which support domestic abuse, controlling and/or hurting
someone, or having sexual activity with someone who did not want/could not refuse
it since the pandemic began.
Almost three quarters of participants (71.8% of 163 people) said “no” to this
question. The majority of those who said “yes” were mainly at home during the
pandemic and had noticed that they were seeing such behaviour more since the
pandemic began. They indicated that most of this content was shared by men.
For the most part, the bystanders who witnessed this took action against it
(73.3% of 45) because they recognised the views were problematic. The main
barrier for those who did not take action was them perceiving that they did not have
the correct skills set.
4.5 Sexist ‘Banter’
Participants were asked if they had noticed or become concerned about general
banter, jokes, videos or statements that are sexist or play on old fashioned gender
roles since the pandemic began.
Just over half of participants
(56%) answered “no” to this
question
Of those that said yes (73 people),
the majority (89.2%) were women
who were mainly at home during
the pandemic (70.3%)
Respondents indicated that most of the problematic statements were made by men.
Generally, bystanders indicated that they had taken action (68.5% of 73 people) in
response to what they had seen by reporting it as they recognised that the opinion of
the individual was inappropriate and felt responsible for doing something about it.
The remaining 31% who did not take action indicated that the barrier for their
behaviour was other people not being concerned or the worry that others would not
support their actions as prosocial bystanders.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
32
5.0 Discussion
5.1 Key Findings
Theadvertisementsforthesurveyspecically
used the phrase ‘domestic abuse’ (see section
3.6) and thus might have signalled to potential
participants that they should have some level
of knowledge about domestic abuse in order to
take part. Given that an important contributor
to survey engagement appears to be the
understanding of DVA (see section 4.1), it may
be that those without knowledge of DVA did not
recognise problematic behaviours and/or their
bystander experiences as being DVA- related,
and therefore did not think that taking the survey
was relevant to them. Thus whilst the survey
questions themselves explored the complex and
multiple behaviours constituting domestic abuse,
recruitmenttothesurveyintherstplacewas
dependent upon the public identifying with the
short descriptor in recruitment advertisements.
These advertisements by their nature cannot be
a long descriptor of the many kinds of behaviours
that constitute DVA. Recruitment of those who
are witnessing DVA but are unaware that it is
DVA they are witnessing remains a fundamental
challenge for this type of research requiring
further exploration.
Women were far more likely than men to take
part in the survey (see section 4.1). There is some
evidence that women are generally more likely
to participate in surveys than men (Smith, 2008).
However, the gendered nature of victimisation,
and the fact that it is predominantly women who
work in the violence against women sector, may
also partly explain why more women than men
lledinthesurvey.
Most participants highlighted that they felt the
pandemic had allowed them to become aware of DVA,
and coercive control and warning signs of DVA were
the most commonly reported behaviours that survey
respondents had witnessed.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
33
Most participants highlighted that they felt the
pandemic had allowed them to become aware
of DVA, and coercive control and warning signs
of DVA were the most commonly reported
behaviours that survey respondents had
witnessed (see section 4.3 and Figure 1). This
ndingaddsweighttoexpertacademicand
practitioner concerns that perpetrators may use
the pandemic to (more) fully control the social
lives and means of correspondence (phone,
computer) of victims (Gulati et al., 2020; Boxall
etal.,2020).Itisapositivendingthatcoercive
control, which became a criminal offence only
relatively recently in 2015, is being noticed by
bystanders.
Interestingly those responding to the survey were
mostly also action takers: although in recruitment
we advertised for experiences of witnessing
DVA behaviours, we also found that that action
takerswerellinginthesurvey(seesection
4.4). Although most respondents had noticed
coercive control or warning signs, only those who
had witnessed sexual abuse or the abuse of a
vulnerable person were certain to take action
against it (see section 4.4). This may be down to
the severity of the behaviour witnessed. Kofman
andGarn(2020)suggestthatbystanderswho
perceive an abusive behaviour to be unlawful or
life-threateningaresignicantlymorelikelyto
ofciallyreportittotheauthorities.
Bystanders were more likely to take action when
told directly by the victim themselves, or when
they witnessed the abuse directly in an intimate
relationship. Relatively few survey respondents
had witnessed the abuse of a man (see section
4.4). There are various explanations for why
this may be so. It is possible that this is due to
bystanders not recognising abuse towards a
male victim as DVA. It is also possible that men
are not offered support by bystanders due to
social norms which perpetuate the idea that DVA
against a man is not “as serious” as DVA against
women (Warburton and Raiolo, 2020). It is also
possible that because DVA is a gendered crime,
men are not being abused as frequently leaving
less scope for bystanders to witness it.
Having knowledge, being able to notice
behaviour as problematic, assuming a sense
ofresponsibility,andbeingcondentinthe
possession of the correct skills are the crucial
steps to being a prosocial bystander and taking
action (Latané and Darley, 1979; Berkowitz,
2009). Interestingly, our data is consistent with
this (see section 4.4). The vast majority took
action.Thendingsillustratethatparticipants
felt responsible and recognised the behaviour as
problematic. Notably the data also showed that
a key barrier to taking action was not recognising
the behaviour as problematic. Further, possessing
the correct skills was revealed to be a crucially
important motivating factor as all participants
who intervened felt they had the skills to do so.
It also appears that a sense of responsibility is
heightened when the victim tells the bystander
themselves, or the bystander witnesses the abuse
directly.
Sense of community appears to be an important
factorininuencingbystanderstotakeaction.
As explored in section 4.4, a heightened sense
of community was linked to several motivational
factors for bystander responses to DVA.
Community action has increased during the
restrictions imposed by COVID-19, which may
have encouraged more prosocial bystander
responses. The link between sense of community
and prosocial bystander behaviours is already
established (Banyard et al., 2020; McMahon et al,
2015; see section 2.2).
Sense of community appears to be an important factor in
inuencing bystanders to take action. Community action has
increased during the restrictions imposed by COVID-19, which
may have encouraged more prosocial bystander responses.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
34
In taking action on DVA, the role of prior
victimisation appears important: being a victim/
survivor oneself is revealed to be a motivator to
taking action (see section 4.4). Those who have
survived DVA themselves can feel a stronger
sense of empathy (from experience) for the
victim. The literature suggests that empathy is
asignicantpredictorofbystanderbehaviours
(Christensen and Harris, 2019; Muralidharan and
Kim, 2019). When a survivor supports a victim,
it can validate the progress they have made
and give them a sense of agency (Gregory et al.,
2016). Further, those who have survived DVA
mayalsobecondentintheirskillsettooffer
assistance to people. Self-perceived possession
of the skills was also true of those working in the
eldwithrelevantDVAtraining,andthosewho
had done training, all of whom were more likely
to take action.
The operation of social norms also appears
importantinourndings(seesection4.4).
When someone else was aware of the abuse,
the bystander was more likely to intervene,
regardless of how often they witnessed it. This
could be attributed to social norms theories in
two ways. Firstly, a consensus that the behaviour
is wrong empowers the bystander to take action
in the knowledge they are going to be supported
and, secondly, peer pressure can make a
bystander do what is socially expected
when someone is in need (Brown and
Messman-Moore, 2010; Fenton et al., 2019). By
contrast, the data also shows that when there
were no other witnesses, bystanders were more
likely to take action after seeing the behaviour
multiple times; the more times they witnessed
it, the more likely they were to do something
(consistent with Rowe, 2018).
Despite participants being motivated to
intervene, being a prosocial bystander had
a substantial negative impact on them as
individuals (see section 4.4). Action may leave
the bystander second guessing their actions
and feeling guilty for not acting sooner. This is
consistent with the literature whereby bystanders
often experience negative psychological impact
from supporting a DVA victim and hearing the
details of the abuse suffered (Gregory et al.,
2016). Those who participated in interviews
explained that they did not feel supported, as
victims and bystanders, which may have fed into
their negative experience of being a bystander.
It appears contradictory that when asked if they
would have liked to do more or do something
differently, most said no, whilst simultaneously
reporting negative effects from their actions.
However,coupledwiththendingthatthe
majoritywouldndbystandertraininguseful,
this contradiction may be explained by the fact
that the wide array of bystander strategies and
options were not known to them and therefore
they had no options to act differently. Further,
bystandertrainingaimstoincreasecondencein
a newly acquired skillset and support for action
which may alleviate some of the negative effects
such as second guessing and doubt about having
done the right thing (see section 2.3).
The bystanders interviewed explained that
they often did not know where to report their
concernsandndingthecorrectadviceonline
can be arduous (see section 4.4). Bystanders
are more likely to report their concerns if they
know how to, if they know they will be believed
and if they know there will be no repercussions
for themselves (Borum, 2013; Rowe, 2018). At
the beginning of the pandemic, the WHO issued
guidelines to policy makers urging them to ensure
members of the public knew what services
were still available to them concerning DVA
(Pearson et al., 2021). However, the impact of the
pandemic was also felt by participants who were
not sure the extent to which they could help the
victim due to government restrictions. This links
with making communities aware of the services
available, making these services easily accessible
and offering reassurance when a bystander
comes forward with concerns (Pearson et al.,
2021; Bradbury-Jones and Isham, 2020).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
35
5.2 Limitations
Firstly, the survey was delivered online through
the web-based platform, Qualtrics. This limited
participants to internet users. It is possible that
data is missing from people who may not use the
internet and social media, or to individuals whose
access to such platforms is limited.
The survey had a large attrition rate, whereby
participantsstoppedllinginthequestionnaire
part way through (see section 4.1). This is
possibly due to the long-time commitment
needed to complete the survey, or could be
attributed to participant fatigue. Furthermore, it
is possible that bystanders had witnessed DVA
prior to the pandemic, and when they reached the
“DVA witnessed during the pandemic” section,
they felt that their experiences were no longer
relevant, and subsequently closed down the
survey. Similarly, participants may not have been
aware that the survey was aimed at people who
had witnessed DVA and as there was no option
for “not witnessed DVA”, they could not continue.
This should be considered in future iterations of
the survey.
When designing the survey, the team were
careful not to use the word “bystander” as they
were not sure members of the public would
know what this means (see section 3.6). Instead,
advertisements were aimed at anyone who had
seen or become concerned about DVA since the
pandemic began. Despite reiterating that the
surveywasspecicallytargetedatthosewho
had seen this in someone else’s relationship, it
is likely that some victims of DVA completed the
survey. Two of the interview participants were
not bystanders, they were victims, and they
thought that the survey had to be worded in such
a way to protect their identities. It is therefore
possible that other victims assumed the same
and completed the survey. Unfortunately, unless
theyhavespecicallystatedthisinoneofthe
open text boxes, there is no way of knowing.
Those who participated in interviews, all
had good knowledge of DVA through their
professional roles or personal experience of
being a survivor themselves. Whilst only a
small number of interviews were conducted,
it suggests that only people with an increased
interest in tackling DVA were willing to share
their experiences further. The survey was lengthy,
and participants were feeling fatigued by the
time they reached the end. It might, therefore,
have been too much of an ask to request them
to contact the research team for additional
information regarding interview, instead of a
simple tick box as part of the survey.
During the interviews, some participants made
comparisons between the UK and Welsh
Governments’ actions during the pandemic and
perpetrator behaviours, namely, the person
having their freedom controlled and being under
restrictions in terms of limiting contact with
family and friends. It is possible that survey
respondents also made this link and did not
distinguish between the two when indicating
what behaviours they had witnessed. This
may contribute to coercive control being the
most reported DVA behaviour (Figure 1) as
participantswerenotspecicallyfocussingon
DVA. For example, as one interview participant
explained;
These ndings offer a preliminary insight into the experiences of bystanders to
DVA during the COVID-19 pandemic in Wales, yet there are some limitations to
this study which need to be taken into account when considering the ndings.
“This isolation period is exactly like being in the abusive relationship.
Not being allowed to go out, not being allowed to see your friends,
having strict conditions on your life. Essentially, the government is
the perpetrator, from a survivors’ perspective” [Interview 2]
"

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
36
This study used an opportunistic and
self-selecting sample. The views of participants
may differ from others who were bystanders to
DVA, or the wider community more generally.
Specialist organisations supporting BAME victims
and wider BAME communities were engaged
during the dissemination of the survey. Despite
this, a large portion of the survey respondents,
andallinterviewees,identiedasWhite
British (see section 4.1). Whilst this is broadly
representative of the Welsh population, the data
may not represent the experiences of those from
other ethnic backgrounds. Likewise, the survey
did not capture any responses from those over
75. How to access potentially harder to reach
BAME and older age groups should be further
explored and improved in future studies.
5.3 Future Actions
This pilot study was situated within the unique
set of circumstances that came about with the
COVID-19 pandemic and the related lockdowns
and social restrictions. Whilst the exact
replication of the conditions for this study may be
difcult,thesurveydesignisnotCOVID-specic,
allowing for replication in non-pandemic contexts.
Further, the learning from developing, delivering
and reporting on this study can be utilised outside
the COVID-19 pandemic because domestic abuse
is an ongoing public health emergency not unique
to the pandemic. The following suggestions are
based upon this.
Policy Options
This study demonstrates that bystanders have
an important role in the primary prevention of
DVA. This should be recognised in VAWDASV
and violence prevention policy. The actions taken
by prosocial bystanders may be an essential
part of tackling DVA at a community level.
Encouraging prosocial behaviour, when safe
to do so, should be a priority both generally
and particularly during a pandemic or other
emergency situations, when services are not as
readily accessible. Policymakers could consider
the use of public awareness campaigns and
training to promote knowledge about DVA, and
prosocial and informed bystander behaviour. This
could help address the barriers to bystanders
taking action and mitigate any negative impact
that the experience may have on the bystander
themselves (see section 4.4).
Lessons from this research suggest that
public-facing bystander campaigns should be
multi-faceted and underpinned by awareness
and knowledge raising of what constitutes DVA
for a public-facing audience. As noted in section
4.4, most participants had witnessed DVA
behaviours within intimate relationships, when
the perpetrator was a man, and the victim was
a woman. Campaigns should emphasise that
abuse can happen in a variety of relationships,
regardless of gender identity, sexuality, age,
ethnicity etc. The array of DVA behaviours should
be made explicit to increase the likelihood of
harmful behaviour recognition; from warning
signs through to coercive control and physical
abuse.
Policy makers should consider how they can
engage different target audiences in knowledge
and awareness raising, and as prosocial
bystanders. In particular, as noted in section
4.1, most participants in this study were women.
Particular attention should be paid to ensuring
men engage in bystander efforts. Awareness
raising campaigns should aim to increase a
sense of responsibility and motivation to act
and therefore be accompanied by the offer of
evidence-based bystander training to enable and
empower bystanders to move (safely) through
the theory of behaviour change (see section 5.1).
The data indicated that a sense of community
wasfoundtobeasignicantpredictorof
bystanders taking prosocial action against DVA
(see section 4.4). Other research has suggested
that community action has increased during
the restrictions imposed by Covid-19 which
may have encouraged more bystander action.
Therefore, policies could aim to nurture, sustain
and further encourage this sense of community
as a contributory factor motivating bystanders to
taking action against DVA.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
37
Practice Options
The research demonstrated that there is a
demand for multifaceted bystander training and
information programmes to empower and upskill
bystanders to take prosocial action, even from
those who are already knowledgeable and have
had training (see section 4.4). Bystander training
programmes must be evidence and
theory-based (Fenton and Mott, 2017). They
should take people through the process of
change including: awareness and recognition
of the gendered nature of DVA in all its forms;
cultural contexts, gender roles and problematic
masculinity underpinning and shoring up DVA;
impacts and empathy; sense of responsibility,
motivationandcondencetoact,andskills
learning.
Social norms theory should be incorporated
in bystander training programmes, materials
and campaigns. For an example of a bystander
training programme for general communities,
see the DVA bystander intervention Active
Bystander Communities (see section 2.3). As
noted in section 4.4, the bystander feeling that
they possess the correct skill set to take action is
essential. A variety of bystander responses could
be incorporated into the campaigns, materials
and bystander training programmes, including
supporting the victim, addressing the perpetrator/
abuser’s behaviour8 and encouraging the use
of services (for example, support services, the
police). For examples of good practice, see
Snowdon et al. (2020). These could be rolled out
across communities to support bystanders as an
essential element in the primary prevention of
DVA, both in and out of a pandemic.
When considering public awareness campaigns
targeted at bystanders, where appropriate,
organisations should make clear which
services they offer that might be of relevance
to bystanders. As discussed in section 5.1,
bystanders are more likely to share their concerns
if they know how to. Clear signposting of relevant
bystander services would allow bystanders to
have increased chance of building knowledge
about what is available. These could be delivered
over an array of platforms, with physical
advertisements, newspapers, online, radio
and television as part of the public awareness
campaign.
As previously highlighted section 1.4, an
increasing number of calls are being recorded
with domestic abuse helplines and the police
from concerned third parties (neighbours,
friends and family). As noted in section 4.4,
bystanders are often negatively impacted by
their experience. Frontline services, including
the police and specialist DVA services, could
consider developing guidance and training for
callhandlersandrst-responderstosupport
bystanders who make contact.
Research Options
This pilot study has tested the methodology,
dissemination and topic area of bystanders
to DVA during the COVID-19 pandemic. The
results are promising. The next step is to revisit
the methodology, and make revisions based on
our learning from implementation and analysis.
Dissemination and survey recruitment should
be amended in future research, to optimise the
recruitment of men, BAME groups, and elderly
people, as discussed in section 5.2. Similarly,
targetingofspecicgroupsworkingwithin
the community (for example, delivery drivers
or community groups) could be improved.
Recruitment should also run for a longer period
of time to optimise uptake and on a larger scale
with a population level sample.
The survey respondents indicated that their
experience of taking action had a negative impact
upon themselves, yet more than half indicated
that there was no further actions they wished
they had taken (see section 4.4). Future research
could explore how these negative impacts could
be mitigated with bystander training programmes
and/or adequate support resources.
Those with a greater sense of community
weresignicantlymorelikelytotakeactionin
response to their concerns (see section 4.4).
Future research should determine what “sense
of community” means to each participant, such
as locality, religion, sports group etc. and how
thissenseofcommunityreectsonthetypesof
behaviour witnessed or the types of action taken.
Further research could also explore the inherent
difcultyinaskingpeopletoparticipate
in research about the DVA they may have
witnessed when they may not recognise what
behaviours constitute DVA (see section 5.1
for discussion on this). This could be achieved
rstlybyincreasedpublicawarenessofwhat
DVA is, and secondly, by alternative methods of
recruitment campaigning which offers more
in–depthdenitionsandexamplesofbehaviours.
8 For a discussion of this point see Fenton et al. (2019).

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
38
Findings from this study suggest that the
circumstances of the pandemic have increased
people’s opportunities to be active bystanders
to DVA behaviours. Participants reported being
more aware of ‘concerning’ behaviours due to
increased time spent at home, coupled with less
‘distraction’ from the norms of regular social and
work life, and a heightened sense of community.
Bystanders also reported that they felt the
circumstances of the pandemic (lockdown,
working from home and social distancing
restrictions) had increased the ability of
perpetrators to further control the victim.
Warning signs of DVA and coercive control were
the most common types of DVA behaviours that
participants had seen or become concerned
about.
The most common action was offering support
to the victim, with the majority of bystanders
offering this. Having received training was a
strong predictor of offering prosocial support
to the victim. Conversely, not having the skills
tonoticeorintervene,wasthemostsignicant
predictor of inaction. Provision of evidence-based
training to bystanders, providing them with the
knowledge,condenceandskillstoidentify
DVA and intervene safely and appropriately,
may mitigate these barriers to taking action.
Bystanders also indicated that having DVA
bystander training would have been useful to
them in guiding them in how to take appropriate
prosocial action.
This study sought to explore the experiences and
behaviours of bystanders to DVA during the COVID-19
pandemic through a mixed methods approach with the
general public, including survey and interviews.
Whilst implemented on a small scale, this study was the
rst of its kind and provides new insights into bystander
experiences during a global pandemic.
6.0 Concluding Comments

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
39
References
Addis, S. & Snowdon, L. (2021) What Works to Prevent Violence Against Women and Girls Domestic Abuse and
Sexual Violence (VAWDASV)? Systematic Evidence Assessment. Wales Violence Prevention Unit, Cardiff.
Assembly UNG. (1979) Convention on the Elimination of All Forms of Discrimination against Women UN Women.
Banyard, V, L. (2015) Toward the next Generation of Bystander Prevention of Sexual and Relationship Violence:
Action Coils to Engage Communities. Switzerland, Springer.
Banyard, V, L., Rizzo, A, J. & Edwards, K. M. (2020) Community Actionists: Understanding Adult Bystanders to
Sexual and Domestic Violence Prevention in Communities. Psychology of Violence, 20(11), pp. 112.
Berg, D.R., Lonsway, K.A. & Fitzgerald, L.F. (1999) Rape Prevention Education for Men: The Effectiveness of
Empathy-induction Techniques. Journal of College Student Development, 30, pp. 219-234.
Berkowitz, A, D. (2003) Applications of Social Norm Theory to Other Health and Social Justice Issues. San
Francisco, Jossey-Bass.
Berkowitz, A. (2009) Response Ability: A Complete Guide to Bystander Intervention. Chicago, Beck.
Borsky, A. E., McDonnell, K., Turner, M. M. & Rimal, R. (2018) Raising a Red Flag on Dating Violence: Evaluation of a
Low-Resource, College-Based Bystander Behavior Intervention Program. Journal of Interpersonal Violence, 33,
pp. 3480-3501.
Borum, R. (2013) Informing Lone-Offender Investigations. Criminology and Public Policy, 12(1), pp. 103-112.
Bouillon-Minois, J., Clinchamps, M. & Dutheil, F. (2020) Coronavirus and Quarantine: Catalysts of Domestic Violence.
Violence Against Women, pp. 1-3.
Boxall, H., Morgan, A. & Brown, R. (2020) The Prevalence of Domestic Violence among Women during the
COVID-19 Pandemic. Australasian Policing, 12(3), pp.38-Bradbury-Jones, C. & Isham, L. (2020) The Pandemic
Paradox: The Consequences of COVID-19 on Domestic Abuse. Journal of Clinical Nursing,
pp. 1-3.
Braun, V. & Clarke, V. (2006) Using Thematic Analysis in Psychology. Qualitative Research in Psychology, 3(2),
pp. 77-101.
Brown, A, L. & Messman-Moore, T, L. (2009) Personal and Percieved Peer Attitudes Supporting Sexual Aggression
as Predictors of Male College Students’ Willingness to Intervene Against Sexual Aggression. Journal of
Interpersonal Violence, 25(3), pp. 503-517.
Campbell, A, M. (2020) An Increasing Risk of Family Violence during the COVID-19 Pandemic: Strengthening
Community Collaborations to Save Lives. Forensic Science International: Reports, pp. 1-3.
Carlyle, K. E., Conley, A. H. & Guidry, J. P. D. (2020) Development and Evaluation of the Red Flag Campaign for the
Primary Prevention of Sexual and Dating Violence on College Campuses. Journal of American College Health,
pp. 1-5.
Christensen, M, C. & Harris, R, J. (2019) Correlates of Bystander Readiness to Help among a Diverse College
Student Population: An Intersectional Perspective. Research in Higher Education, 60(8), pp. 1195-1226.
Coker, A. L., Bush, H. M., Brancato, C. J., Clear, E. R. & Recktenwald, E. A. (2019) Bystander Program Effectiveness
to Reduce Violence Acceptance: RCT in High Schools. Journal of Family Violence, 34, pp. 153-164.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
40
Edwards, K, M., Banyard, V, L., Sessarego, S. N., Waterman, E. A., Mitchell, K. J. & Chang, H. (2019) Evaluation of a
Bystander-Focused Interpersonal Violence Prevention Program with High School Students. Prevention Science, 20,
pp. 488-498.
Europe Co. (2011) The Istanbul Convention. Council of Europe.
Fenton, R., Mott, H., McCartan, K. & Rumney, P. (2016) A Review of Evidence for Bystander Intervention to prevent
Sexual and Domestic Violence in Universities. Available at https://assets.publishing.service.gov.uk/government/
uploads/system/uploads/attachment_data/le/515634/Evidence_review_bystander_intervention_to_prevent_
sexual_and_domestic_violence_in_universities_11April2016.pdf [Accessed March 2021].
Fenton, R. & Mott, H. (2018) The Intervention Initiative: Theoretical Underpinnings, Development and
Implementation. Gender Based Violence in University Communities: Policies, Prevention and Educational Initiatives,
pp. 169-188.
Fenton, R. & Mott, H. (2019) Evaluation of the Intervention Initiative: A Bystander Intervention Program to Prevent
Violence Against Women in Universities. University of Exeter.
Fenton, R., Jones, C., Moss, S. & Cooke, K. (2019) The Challenges of Developing and Implementing a Bystander
Intervention for the Prevention of Domestic Violence and Abuse in UK Communities. Journal of Gender Based
Violence, 3(2), pp 215-231.
Fleming, M. & Wiersma-Mosley, J, D. (2015) The Role of Alcohol Consumption Patterns and Pro-Social Bystander
Interventions in Contexts of Gender Violence. Violence Against Women, 21(10), pp. 1259-1283.
Gainsbury, A. N., Fenton, R. A. & Jones, C. A. (2020) From Campus to Communities: Evaluation of the First UK-
Based Bystander Programme for the Prevention of Domestic Violence and Abuse in General Communities. BMC
Public Health, 20, pp. 674.
GovernmentH.(2016)StrategytoEndViolenceAgainstWomenandgirls:2016to2020.HomeOfce:HM
Government.
Green, D, P., Wilke, A, M. & Cooper, J. (2020) Countering Violence Against Women by Encouraging Disclosure: A
Mass Media Experiment in Rural Uganda. Comparative Political Studies, 53(14), pp. 2283-2320.
Gregory, A, C., Williamson, E. & Feder, G. (2016) The Impact on Informal Supporters of Domestic Violence Survivors:
A Systematic Literature Review. Trauma Violence and Abuse, pp. 1-19.
Gulati, S., Sarai, S, K. & Lippman, S. (2020) COVID-19 and Domestic Violence. Available at COVID-19 and domestic
violence – Internet and Psychiatry [Accessed March 2021].
Hester, M (2013) Who Does What to Whom? Gender and Domestic Violence Perpetrators in English Police
Records. European Journal of Criminology, 10(5), pp. 623-637.
Hester, M. & Lilley, S. J. (2014) Preventing Violence Against Women: Article 12 of the Istanbul Convention.
Strasbourg: Council of Europe.
Ivandic, R., Kirchmaier, T, & Linton, B. (2020) Changing Patterns of Domestic Abuse during COVID-19 Lockdown.
SSRN, p. 50.
Jouriles, E. N., Krauss, A., Vu, N. L., Banyard, V. L. & Mcdonald, R. (2018) Bystander Programs Addressing Sexual
Violence on College Campuses: A Systematic Review and Meta-analysis of Program Outcomes and Delivery
Methods. Journal of American College Health, 66, pp. 457-466.
Jurasz, O. (2018) Violence Against Women, Domestic Abuse and Sexual Violence (Wales) Act 2015.
Kaukinen, C. (2020) When Stay-At-Home Orders Leave Victims Unsafe at Home: Exploring the Risk and
Consequences of Intimate Partner Violence during the COVID-19 Pandemic. American Journal of Criminal Justice,
45, pp. 668-679.
Kettrey, H. H. & Marx, R. A. (2019a) Does the Gendered Approach of Bystander Programs Matter in the Prevention
of Sexual Assault Among Adolescents and College Students? A Systematic Review and Meta-Analysis. Archives of
Sexual Behaviour, 48, pp. 2037-2053.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
41
Kettrey, H. H. & Marx, R. A. (2019b) The Effects of Bystander Programs on the Prevention of Sexual Assault across
the College Years: A Systematic Review and Meta-analysis. Journal of Youth and Adolescence, 48, pp. 212-227.
Kofman,Y,B.Garn,D,R.(2020)HomeisNotAlwaysaHaven:TheDomesticViolenceCrisisamidtheCOVID-19
Pandemic. Psychological Trauma: Theory, Research, Practice and Policy, 12(81), pp. 5199-5201.
Kovalenko, A. G., Abraham, C., Graham-Rowe, E., Levine, M. & O'Dwyer, S. (2020) What Works in Violence
Prevention Among Young People?: A Systematic Review of Reviews. Trauma Violence Abuse, 1524838020939130.
McCarry, M., Larkins, C., Berry, V., Radford, L. & Stanley, N. (2018) The Potential for Co-production in Developing
Violence against Women Services in Wales. Social Policy and Society, 17, pp. 193-208.
McMahon, S. (2015) Call for Research on Bystander Intervention to Prevent Sexual Violence: The Role of Campus
Environments. American Journal of Community Psychology, 55, pp. 472-489.
Mujal, G. N., Taylor, M. E., Fry, J. L., Gochez-Kerr, T. H. & Weaver, N. L. (2019) A Systematic Review of Bystander
Interventions for the Prevention of Sexual Violence. Trauma Violence Abuse, 1524838019849587.
Muralidharan, S. & Kim, E. (2019) ‘I Owe My Life to the Kind Individual who called the Helpline’: Exploring the
Interplay of Self-Views and Social Distance in Narrative Advertising on Domestic Violence Prevention in India and
the United States. International Journal of Advertising, 38(5), pp.704-730.
NICE (2014) Domestic Violence and Abuse: Multiagency Working. Available at Introduction | Domestic violence and
abuse: multi-agency working | Guidance | NICE [Accessed March 2021]
OfceforNationalStatistics.(2020a)Domestic Abuse in England and Wales Overview. Available at Domestic
abuseinEnglandandWalesoverview-OfceforNationalStatistics(ons.gov.uk)[AccessedMarch2021].
OfceforNationalStatistics.(2020b)DomesticAbusePrevalenceandTrends,EnglandandWales:YearEnding
March2020.AvailableatDomesticabuseprevalenceandtrends,EnglandandWales-OfceforNational
Statistics (ons.gov.uk) [Accessed March 2021]
OfceforNationalStatistics.(2020c)Domestic Abuse during the Coronavirus (COVID-19) Pandemic, England
and Wales: November 2020. Available at https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/
articles/domesticabuseduringthecoronaviruscovid19pandemicenglandandwales/november2020
[Accessed March 2021]
OfceforNationalStatistics.(2019)Domestic Abuse in England and Wales. Available at Domestic abuse in
EnglandandWales:November2019-OfceforNationalStatistics(ons.gov.uk) [Accessed March 2021].
Oliver, R., Alexander, B., Roe, S. & Wlasny, M. (2019) The Economic and Social Costs of Domestic Abuse. Home
Ofce.
Pearson, I., Butler, N., Yelgezekova, Z., Nihlen, A., Aguirre, I, Y., Quigg, Z. & Stockl, H. (2021) Emerging Responses
Implemented to Prevent and Respond to Violence against Women and Children in WHO European Member States
during the COVID-19 Pandemic: A Scoping Review of Online Media Reports. BMJ Open, 11(4), e045872.
Price, J., Quarmby, S., Johnson, C. & Bristow, D. (2020) Domestic Abuse Interventions in Wales. Wales Centre for
Public Policy.
Riggs, D.W., Taylor, N., Fraser, H., Donovan, C., & Signal, T. (2021) The Link Between Domestic Violence and Abuse
and Animal Cruelty in the Intimate Relationships of People of Diverse Genders and/or Sexualities: A Binational
Study. Journal of Interpersonal Violence, 36, pp 5-6.
Robinson,A,L.(2017)SerialDomesticAbuseinWales:AnExploratoryStudyintoitsDenition,Prevalence,
Correlates and Managements. Victims and Offenders, 12(5), pp. 642-662.
Rowe, M. (2018) Fostering Constructive Action by Peers and Bystanders in Organisations and Communities.
Negotiation Journal, 34(2), pp. 137-163.

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
42
Sacco, M, A., Caputo, F., Ricci, P., Sicilia, F., Aloe, L, D., Bonetta, F., Cordasco, F., Scalise, C., Cacciatore, G., Zibetti, A.,
Gratteri, S. & Aquila, I. (2020) The Impact of the COVID-19 Pandemic on Domestic Violence: The Dark Side of Home
Isolation during Quarantine. Medico-Legal Journal, 88(2), pp. 71-73.
Smith, W, G. (2008) Does Gender Inuence Online Survey Participation? Available at ERIC - ED501717 - Does
GenderInuenceOnlineSurveyParticipation?ARecord-LinkageAnalysisofUniversityFacultyOnlineSurvey
Response Behavior, Online Submission, 2008-Jun [Accessed July 2021].
Snowdon, L, C., Barton, E, R., Newbury, A., Parry, B., Bellis, M, A. & Hopkins, J, C. (2020) Addressing the “Shadow
Pandemic” through a Public Health Approach to Violence Prevention. Journal of Community Safety and Well-Being,
5(2), pp. 60-65.
Snowdon, L, C., Gainsbury, A, Fenton, R, Kovalenko, A., Kam, R., Sims, N., Bath, R., O’Moore, E., Womack, J.
& Heasman, M. (2020) Bystander Intervention to Prevent Intimate Partner and Sexual Violence. Available at
Bystander interventions to prevent intimate partner and sexual violence https://www.gov.uk/government/
publications/interventions-to-prevent-intimate-partner-and-sexual-violence/bystander-interventions-to-prevent-
intimate-partner-and-sexual-violence-summary [Accessed June 2021].
Speed, A., Thomson, C. & Richardson, K. (2020) Stay Home, Stay Safe, Save Lives? An Analysis of the Impact of
COVID-19 on the Ability of Victims of Gender-Based Violence to Access Justice. The Journal of Criminal Law,
pp. 1-36.
Storer,H.L.,Casey,E.&Herrenkohl,T.(2016)EfcacyofBystanderProgramstoPreventDatingAbuseAmong
Youth and Young Adults:A Review of the Literature. Trauma, Violence, & Abuse, 17, pp. 256-269.
Taylor, E., Banyard, V., Grych, J. & Hamby, S. (2016) Not All Behind Closed Doors: Examining Bystander Involvement
in Intimate Partner Violence. Journal of Interpersonal Violence, pp. 1-21.
UnitedNationsOfceofDrugsandCrime.(2019)Global Study on Homicide. Available at Global study on homicide
(unodc.org) [Accessed March 2021].
UN Women. (2020) Violence Against Women and Girls: The Shadow Pandemic. Available at Violence against
women and girls: the shadow pandemic | UN Women – Headquarters [Accessed March 2021].
Wales Violence Prevention Unit (2020). Monthly Monitoring Report. Cardiff, Wales VPU.
Warburton, E. & Raniolo, G. (2020) Domestic Abuse during COVID-19: What about the Boys? Psychiatry Research,
pp. 291.
Weitzman, A., Cowan, S. & Walsh, K. (2020) Bystander Interventions of Behalf of Sexual Assault and Intimate
Partner Violence Victims. Journal of Interpersonal Violence, 35(7), pp. 1694-1718.
Welsh Government. (2005) Tackling Domestic Abuse: The All Wales National Strategy.
Welsh Government. (2010) Right to be safe Strategy.
Welsh Government. (2012) Consultation on Legislation to End Violence against Women, Domestic Abuse and
Sexual Violence (Wales).
Welsh Government. (2015) The Violence Against Women, Domestic Abuse and Sexual Violence (Wales) Act 2015.
Available at guidance-for-local-strategies.pdf (gov.wales) [Acessed March 2021].
Welsh Government. (2016) National Strategy on Violence against Women, Domestic Abuse and Sexual Violence-
2016-2021. Available at national-strategy-2016-to-2021.pdf (gov.wales) [Accessed March 2021].
Welsh Government. (2020) Ethnicity by Region. Available at Ethnicity by Region (gov.wales)
[Accessed 1st April 2021].
Welsh Women’s Aid. (2020) A Perfect Storm. Available at A-Perfect-Storm-August-2020-1.pdf
(womensaid.org.uk) [Accessed March 2021].
World Health Organisation. (2021) Violence Against Women. Available at https://www.who.int/news-room/
fact-sheets/detail/violence-against-women#:~:text=These%20forms%20of%20violence%20can,experience%20
depression%20and%20problem%20drinking [Accessed March 2021].

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
43
Appendices
A: Participant Demographics
Table 5 Participant Demographics
Demographic Number of Participants
Gender
• Man
• Woman
• 28
• 158
Age
• 18-34
• 35-44
• 45 +
• 69
• 54
• 63
Ethnicity
• White British/Irish
• Arab/Asian/African/Caribbean
• 178
• 8
Occupation
• Usually at home9
• Health and social care
• Education
• Local authority or government and
key Public services
• Industry and other tertiary jobs
• 30
• 41
• 31
• 39
• 45
Status during lockdown
• Primarily at home
• Primarily away from home
• 141
• 45
Knowledge of DVA
• Extremely/very
• Moderately/slightly
• Not at all
• 119
• 66
• 1
Knowledge of DVA law
• Extremely/very
• Moderately/slightly
• Not at all
• 71
• 98
• 17
9 Usually at home- retired, stay at home parent, unemployed

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
44
Appendices
B: Data Tables for DVA Witnessed During Pandemic
Table 6 Frequency of behaviours witnessed
Behaviour Witnessed Frequency
Warning signs 132
Coercive control 168
LGBTQI+ 25
Vulnerability 40
Threats 81
Physical 83
Sexual 35

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
45
Table 7 Chi-squared ndings for behaviours witnessed and participant demographics
Behaviours Witnessed
Warning
Signs
Coercive
Control LGBTQI+ Vulnerability Threats Physical Sexual
Man 15
(53.6%)
24
(85.7%)
4
(14.3%)
5
(17.9%)
16
(57.1%)
10
(35.7%)
2
(7.1%)
Woman 117
(74.1%)
144
(91.1%)
21
(13.3%)
35
(22.2%)
65
(41.1%)
73
(46.2%)
33
(20.9%)
X2(1, N=186)
= 4.842
(1, N=186)
= .801
(1,N=186)
= .020
(1, N=186) =
.260
(1, N=186)
= 2.478
(1, N=186)
= 1.059
(1, N=186)
= 2.941
p 0.028 0.371 0.887 0.610 0.115 0.303 0.086
18-34 52
(39.4%)
66
(39.3%)
7
(28%)
11
(27.5%)
24
(29.6%)
29
(34.9%)
13
(37.2%)
35-44 39
(29.5%)
49
(29.2%)
5
(20%)
12
(30%)
25
(30.9%)
25
(30.2%)
11
(31.4%)
45+ 41
(31.1%)
53
(31.5%)
13
(52%)
17
(42.5%)
32
(39.5%)
29
(34.9%)
11
(31.4%)
X2(2, N=186)
= 1.748
(2, N=186)
= 5.020
(2, N=186)
= 4.258
(2, N=186) =
2.402
(2, N=186)
= 3.668
(2, N=186)
= .300
(2, N=186)
= .161
P 0.417 0.081 0.119 0.301 0.160 0.861 0.923
GenderAge

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
46
Behaviours Witnessed
Warning
Signs
Coercive
Control LGBTQI+ Vulnerability Threats Physical Sexual
Usually at
home
19
(63.3%)
27
(90%)
3
(10%)
4
(13.3%)
15
(50%)
13
(43.3%)
4
(13.3%)
Health
and social
care
19
(63.3%)
27
(90%)
3
(10%)
4
(13.3%)
15
(50%)
13
(43.3%)
4
(13.3%)
Government
or public
services
26
(66.7%)
27
(90%)
3
(7.7%)
5
(12.8%)
17
(43.6%)
19
(48.7%)
7
(17.9%)
Education 25
(80.6%)
31
(100%)
1
(3.2%)
7
(22.6%)
11
(35.5%)
15
(48.4%)
5
(16.1%)
Industry or
tertiary
32
(71.1%)
37
(82.2%)
7
(15.6%)
16
(35.6%)
19
(42.2%)
16
(35.6%)
9
(20%)
X2(4, N=186)
= 2.705
(4, N=186)
= 6.879
(4, N=186)
= 10.683
(4, N=186) =
8.310
(4, N=186)
= 1.490
(4, N=186)
= 2.247
(4, N=186)
= 1.631
P 0.608 0.142 0.03 0.081 0.828 0.691 0.803
Mainly at
home
103
(73%)
129
(91.5%)
19
(13.5%)
31
(22%)
64
(45.4%)
59
(41.8%)
26
(18.4%)
Mainly
away from
home
39
(64.4%) 39
(86.7%)
6
(13.3%)
9
(20%)
17
(37.8%)
24
(53.3%)
9
(20%)
X2(1, N=186)
= 1.226
(1, N=186)
= .908
(1, N=186)
= .001
(1, N=186) =
.080
(1, N=186)
= .804
(1, N=186)
= 1.822
(1, N=186)
= .054
P 0.268 0.341 0.981 0.778 0.37 0.177 0.816
Occupation Lockdown Status

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
47
Took action (yes)
Gender
Man 24 (85.7%)
Woman 140 (88.6%)
X2(1, N=186) = .191
P 0.662
Age
18-34 56 (81.2%)
35-44 48 (88.9%)
45+ 60 (95.2%)
X2(2, N=186) = 6.296
P 0.043
Occupation
Usually at home 26 (86.7%)
Health and social
care 36 (87.8%)
Government or
public services 38 (97.4%)
Education 25 (80.6%)
Industry or tertiary 39 (86.7%)
X2(4, N=186) = 5.062
P 0.281
Lockdown status
Mainly at home 124 (87.9%)
Mainly away from
home 40 (88.9%)
X2(1, N=186) = .029
P 0.864
Table 8 Chi-squared results for participant demographics and taking action against DVA

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
48
Took action (yes)
Feel connected to
community
Agree 75 (89.3%)
Neither 39 (88.6%)
Disagree 50 (86.2%)
X2(2, N=186) = .324
P 0.851
Want to help
members of my
community
Agree 133 (89.3%)
Neither 27 (81.8%)
Disagree 4 (100%)
X2(2, N=186) = 1.984
P 0.371
Domestic abuse
awareness a part of
professional role
Yes 80 (92%)
No 84 (84.8%)
X2(1, N=186) = 2.242
P 0.134
Have you done
DVA training in the
past 5 years?
Increased 84 (94.4%)
Stayed the same 80 (82.5%)
X2(1, N=186) = 6.311
P 0.012
Do you know
someone who has
experienced DVA?
Yes 157 (99.7%)
No 7 (77.8%)
X2(1, N=186) = .980
P 0.322

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
49
Took action (yes)
Since the
pandemic began,
my awareness
of DVA has…
Increased 98 (92.5%)
Stayed the same 66 (82.5%)
X2(1, N=186) = 4.330
P 0.037
How knowledgeable
are you about DVA?
Extremely 103 (86.6%)
Moderately 60 (90.9%)
Not at all 1 (100%)
X2(2, N=186) = .907
P 0.635
How knowledgeable
are you about the
laws relating to DVA?
Extremely 66 (93%)
Moderately 84 (85.7%)
Not at all 14 (82.4%)
X2(2, N=186) = 2.679
P 0.262

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
50
Took action (yes)
Behaviour Witnessed
Warning signs 106 (86.2%)
Coercive control 11 (73.3%)
Vulnerability 6 (100%)
Threats 13 (100%)
Physical 23 (92%)
Sexual 5 (100%)
X2(5, N=186) = 6.931
P 0.226
Frequency the
behaviour was
witnessed
1-4 times 76 (46.3%)
5+ times 88 (53.7%)
X2(1, N=186) = .104
P 0.747
Relationship
between victim and
perpetrator
Family members 32 (84.2%)
Intimate or ex
partners 125 (89.9%)
Unsure 7 (77.8%)
X2(2, N=186) = 1.915
P 0.384
Gender of perpetrator
Man 131 (90.3%)
Woman 31 (88.6%)
Unsure 2 (33.3%)
X2(2, N=186) = 17.963
P 0.000
Table 9 Chi-squared results for details about DVA and taking action

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
51
Took action (yes)
Gender of victim
Man 25 (89.3%)
Woman 137 (89.5%)
Unsure 2 (40%)
X2(2, N=186) = 11.434
P 0.003
How did you initially
come to be witness/
know about the
behaviour?
Physically in person 65 (86.7%)
Told by victim 64 (95.5%)
Told by someone
else 21 (91.3%)
Don’t want to
answer 9 (60%)
Online 5 (83.3%)
X2(4, N=186) = 15.400
P 0.004
Did anyone else see
the behaviour?
Yes 102 (96.2%)
No 34 (87.2%)
Unsure 28 (68.3%)
X2(2, N=186) = 22.167
P 0.000
Relationship to victim
Family 40 (90.9%)
Friend 44 (95.7%)
Acquaintance 29 (82.9%)
Part of a
community group 41 (82%)
Stranger 10 (90.9%)
X2(4, N=186) = 5.637
P .228

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
52
Behaviours chosen to give details about Chi-squared data
Coercive Warning
signs Vulnerability Threats Physical Sexual X2P
Actions taken
Looked for
more info (yes)
49
(40.2%)
4
(26.7%)
3
(50%)
5
(38.5%)
12
(48%)
4
(80%) (5, N=186) = 5.168 0.396
Unofciallyshared
concerns (yes) 53 27
(90%)
3
(10%)
4
(13.3%)
15
(50%)
13
(43.3%) (5, N=186) = 10.207 0.070
Ofciallyshared
concerns (yes)
31
(25.4%)
0
(0%)
2
(33.3%)
10
(76.9%)
9
(36%)
1
(20%) (5, N=186) = 22.448 0.000
Signalled
disapproval or
distracted (yes)
28
(23%)
2
(13.3%)
3
(50%)
2
(15.4%)
10
(40%)
1
(20%) (5, N=186) = 7.114 0.212
Victim support (yes) 62
(50.8%)
9
(60%)
1
(16.7%)
8
(61.5%)
21
(84%)
3
(60%) (5, N=186) = 13.336 0.020
Other (yes) 10
(8.2%)
1
(6.7%)
1
(16.7%)
2
(15.4%)
4
(16%)
0
(0%) (5, N=186) = 2.960 0.706
Table 10 Chi-squared results for actions, motivations and barriers for each DVA behaviour witnessed

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
53
Behaviours chosen to give details about Chi-squared data
Coercive Warning
signs Vulnerability Threats Physical Sexual X2P
Motivations
Recognising (yes) 72
(69.2%)
7
(63.6%)
5
(83.3%)
9
(69.2%)
20
(87%)
5
(100%) (5, N=162) = 5.756 0.331
Feeling
responsible (yes)
72
(68.6%)
9
(81.8%)
5
(83.3%)
11
(84.6%)
21
(91.3%)
4
(80%) (5, N=163) = 6.748 0.24
Correct skills (yes) 33
(31.4%)
3
(27.3%)
3
(50%)
5
(38.5%)
15
(65.2%)
1
(20%) (5, N=163) = 10.788 0.056
Personal reasons
(yes)
44
(41.9%)
7
(63.6%)
4
(66.7%)
7
(53.8%)
16
(69.6%)
2
(40%) (5, N=163) = 7.983 0.157
No reason
provided (yes)
18
(17%)
0
(0%)
1
(16.7%)
0
(0%)
1
(4.3%)
0
(0%) (5, N=164) = 7.730 0.172

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
54
Behaviours chosen to give details about Chi-squared data
Coercive Warning
signs Vulnerability Threats Physical Sexual X2P
Barriers
Not noticing (yes) 6
(31.6%)
0
(0%)
0
(0%)
0
(0%)
0
(0%)
0
(0%) (2, N=25) = 2.493 0.287
Nocondence(yes) 2
(10.5)
0
(0%)
0
(0%)
0
(0%)
1
(50%)
0
(0%) (2, N=25) = 3.319 0.19
Lacking skills (yes) 4
(21.1%)
1
(25%)
0
(0%)
0
(0%)
1
(50%)
0
(0%) (2, N=25) = .834 0.659
Fear of
retaliation (yes)
3
(15.8%)
0
(0%)
0
(0%)
0
(0%)
1
(50%)
0
(0%) (2, N=25) = 2.483 0.289
No motivation(yes) 1
(5.3%)
1
(25%)
0
(0%)
0
(0%)
0
(0%)
0
(0%) (2, N=25) = 1.938 0.379
Pluralistic
ignorance (yes)
2
(10.5%)
0
(0%)
0
(0%)
0
(0%)
0
(0%)
0
(0%) (2, N=25) = .686 0.709
Victim blaming (yes) 1
(5.3%)
0
(0%)
0
(0%)
0
(0%)
0
(0%)
0
(0%) (2, N=25) = .329 0.848
No response given
(yes)
11
(52.6%)
3
(75%)
0
(0%)
0
(0%)
0
(0%)
0
(0%) (2, N=25) = 3.017 0.221

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
55
Motivations
Recognising
the situation
as problematic
Feeling
responsible to
do something
Possessing
the right
skills
Personal
reasons
No reason
provided
DVA
Knowledge
Ver y 75 (72.8%) 78 (75.7%) 45 (43.7%) 53 (51.5%) 11 (10.7%)
Slightly 42 (72.4%) 43 (72.9%) 15 (25.4%) 26 (44.1%) 9 (15%)
Not at all 1 (100%) 1 (100%) 0 (0%) 1 (100%) 0 (0%)
X2(2, N=162) =
.378
(2, N=163) =
.500
(2, N=163) =
5.967
(2, N=163) =
1.863
(2, N=163) =
.801
P 0.828 0.779 0.051 0.394 0.67
DVA Law
Knowledge
Ver y 48 (73.8%) 49 (74.2%) 38 (57.6%) 36 (54.5%) 5 (7.6%)
Slightly 58 (69.9%) 60 (72.3%) 19 (22.9%) 37 (44.6%) 15 (17.9%)
Not at all 12 (85.7%) 13 (92.9%) 3 (21.4%) 7 (50%) 0 (0%)
X2(2, N=162) =
1.574
(2, N=163) =
2.713
(2, N=163) =
20.572
(2, N=163) =
1.467
(2, N=164) =
5.775
P 0.455 0.258 0.000 0.48 0.056
Table 11 Chi-squared results for participant demographics and motivations for taking action

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
56
Motivations
Recognising
the situation
as problematic
Feeling
responsible to
do something
Possessing
the right
skills
Personal
reasons
No reason
provided
Attended DVA
training in the
past 5 years
Yes 63 (75%) 63 (75%) 43 (51.2%) 38 (45.2%) 8 (9.5%)
No 55 (70.5%) 59 (74.7%) 17 (21.5%) 42 (53.2%) 12 (15%)
X2(1, N=162) =
.412
(1, N=163) =
.002
(1, N=163) =
15.409
(1, N=163) =
1.023
(1, N=164) =
1.148
P 0.521 0.963 0.000 0.312 0.284
Feel connected
to the
community
Agree 59 (78.7%) 63 (84%) 37 (49.3%) 35 (46.7%) 3 (4%)
Neither 30 (76.9%) 36 (66.7%) 13 (33.3%) 18 (46.2%) 6 (15.4%)
Disagree 29 (60.4%) 33 (67.3%) 10 (20.4%) 27 (55.1%) 11 (22%)
X2(2, N=162) =
5.360
(2, N=163) =
6.188
(2, N=163) =
10.927
(2, N=163) =
1.019
(2, N=164) =
9.564
P 0.069 0.045 0.004 0.601 0.008

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
57
Motivations
Recognising
the situation
as problematic
Feeling
responsible to
do something
Possessing
the right
skills
Personal
reasons
No reason
provided
I want to help
members of
my community
Agree 103 (78%) 104 (78.8%) 55 (41.7%) 70 (53%) 12 (9%)
Neither 13 (50%) 16 (59.3%) 4 (14.8%) 8 (29.6%) 7 (25.9%)
Disagree 2 (50%) 2 (50%) 1 (25%) 2 (50%) 1 (25%)
X2(2, N=162) =
9.708
(2, N=163) =
5.885
(2, N=163) =
7.194
(2, N=163) =
4.913
(2, N=164) =
6.617
P 0.008 0.053 0.027 0.086 0.037
DVA
awareness a
part of
professional
role
Yes 59 (74.7%) 59 (73.8%) 40 (50%) 36 (45%) 9 (11.3%)
No 59 (71.7%) 63 (75.9%) 20 (24.1%) 44 (53%) 11 (13.1%)
X2(1, N=162) =
.265
(1, N=163) =
.100
(1, N=163) =
11.751
(1, N=163) =
1.046
(1, N=164) =
.130
P 0.607 0.751 0.001 0.306 0.718

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
58
Barriers
Noticing
(yes)
Condence
(yes)
Lacking
correct skills
(yes)
Fear of
retaliation
(yes)
No
Motivations
(yes)
Pluralistic
ignorance
(yes)
Victim
blaming
(yes)
No response
provided
DVA
knowledge
Very 3 (17.6%) 2 (11.8%) 3 (17.6%) 3 (17.6%) 1 (5.9%) 0 (0%) 0 (0%) 10 (58.8%)
Slightly 3 (37.5%) 1 (12.5%) 3 (37.5%) 1 (12.5%) 1 (12.5%) 2 (25%) 1 (12.5%) 3 (37.5%)
Not at all 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
X2(1, N=25) =
1.176
(1, N=25) =
.003
(1, N=25) =
1.176
(1, N=25) =
.107
(1, N=25)=
.324
(1, N=25) =
4.620
(1, N=25) =
2.214
(1, N=25) =
.991
P 0.278 0.958 0.278 0.743 0.569 0.032 0.137 0.319
Law
knowledge
Very 0 (0%) 0 (0%) 1 (20%) 0 (0%) 1 (20%) 0 (0%) 0 (0%) 4 (80%)
Slightly 5 (29.4%) 3 (17.6%) 5 (29.4%) 3 (17.6%) 1 (5.9%) 2 (11.8%) 1 (5.9%) 8 (47.1%)
Not at all 1 (33.3%) 0 (0%) 0 (0%) 1 (33.3%) 0 (0%) 0 (0%) 0 (0%) 1 (33.3%)
X2(2, N=25) =
1.995
(2, N=25) =
1.604
(2, N=25) =
1.264
(2, N=25) =
1.657
(2, N=25) =
1.343
(2, N=25) =
1.023
(2, N=25) =
.490
(2, N=25) =
2.156
P 0.369 0.448 0.531 0.437 0.511 0.600 0.783 0.34
Table 12 Chi-squared results for participant demographics and barriers to taking action

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
59
Barriers
Noticing
(yes)
Condence
(yes)
Lacking
correct skills
(yes)
Fear of
retaliation
(yes)
No
Motivations
(yes)
Pluralistic
ignorance
(yes)
Victim
blaming
(yes)
No response
provided
Training
Yes 1 (16.7%) 0 (0%) 1 (16.7%) 1 (16.7%) 0 (0%) 0 (0%) 0 (0%) 4 (66.7%)
No 5 (26.3%) 3 (15.8%) 5 (26.3%) 3 (15.8%) 2 (10.5%) 2 (10.5%) 1 (5.3%) 9 (47.4%)
X2(1, N=25) =
.233
(1, N=25) =
1.077
(1, N=25) =
.233
(1, N=25) =
.003
(1, N=25) =
.686
(1, N=25) =
.686
(1, N=25) =
.329
(1, N=25) =
.680
P 0.629 0.299 0.629 0.959 0.407 0.407 0.566 0.409
Professional
role
Yes 2 (22.2%) 0 (0%) 2 (22.2%) 2 (22.2%) 0 (0%) 0 (0%) 0 (0%) 6 (66.7%)
No 4 (25%) 3 (18.8%) 4 (25%) 2 (12.5%) 2 (12.5%) 2 (12.5%) 1 (6.3%) 7 (43.8%)
X2(1, N=25) =
.024
(1, N=25) =
1.918
(1, N=25) =
.024
(1, N=25) =
.405
(1, N=25) =
1.223
(1, N=25) =
1.223
(1, N=250) =
.586
(1, N=25) =
1.212
P 0.876 0.166 0.876 0.524 0.269 0.269 0.444 0.271

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
60
Barriers
Noticing
(yes)
Condence
(yes)
Lacking
correct skills
(yes)
Fear of
retaliation
(yes)
No
Motivations
(yes)
Pluralistic
ignorance
(yes)
Victim
blaming
(yes)
No response
provided
Connected to
community
Agree 4 (44.4%) 1 (11.1%) 2 (22.2%) 1 (11.1%) 1 (11.1%) 1 (11.1%) 0 (0%) 4 (44.4%)
Neither agree
nor disagree 2 (33.3%) 1 (16.7%) 3 (50%) 3 (50%) 1 (16.7%) 0 (0%) 1 (16.7%) 0 (0%)
Disagree 0 (0%) 1 (10%) 1 (10%) 0 (0%) 0 (0%) 1 (10%) 0 (0%) 9 (90%)
X2(2, N=25) =
5.507
(2, N=25) =
.168
(2, N=25) =
3.314
(2, N=25) =
7.226
(2, N=25) =
1.600
(2, N=25) =
.694
(2, N=25) =
3.299
(2, N=25) =
12.491
P 0.064 0.919 0.191 0.027 0.449 0.707 0.192 0.002
Help members
of community
Agree 4 (21.1%) 2 (10.5%) 4 (21.1%) 3 (15.8%) 2 (10.5%) 2 (10.5%) 0 (0%) 11 (57.9%)
Neither agree
nor disagree 2 (33.3%) 1 (16.7%) 2 (33.3%) 1 (16.7%) 0 (0%) 0 (0%) 1 (16.7%) 2 (33.3%)
X2(1, N=25) =
.377
(1, N=25) =
.163
(1, N=25) =
.377
(1, N=25) =
.003
(1, N=25) =
.686
(1, N=25) =
.686
(1, N=25) =
3.299
(1, N=25) =
1.102
P 0.539 0.687 0.539 0.959 0.407 0.407 0.069 0.294

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
61
Witnessed Sexist Banter
Yes No
Gender
Man 7 (28%) 18 (72%)
Woman 66 (46.8%) 75 (53.2%)
X2(1, N=166) = 3.049
P 0.081
Age
18-34 35 (55.6%) 28 (44.4%)
35-44 21 (44.7%) 26 (55.3%)
45+ 17 (30.4%) 39 (69.6%)
X2(2, N=166) = 7.654
P 0.022
Lockdown Status
Mainly at home 52 (40.3%) 77 (59.7%)
Mainly away from
home 21 (56.8%) 16 (43.2%)
X2(1, N=166) = 3.157
P 0.076
Table 13 Chi-squared results for participant demographics and witnessing sexist banter
C: Data Tables for Sexist Banter

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
62
Frequency
Gender of the person
who shared this
Man 52
Woman 9
Genderuid/
non-binary 1
Unsure 11
Took action?
Yes 50
No 23
Action Taken
Reported the post or
shared disapproval
44 (yes)
29 (no)
Shared concerns with
others
18 (yes)
55 (no)
Looked for more
information
8 (yes)
65 (no)
Took an alternative
form of action
14 (yes)
59 (no)
Motivation for
taking action
Recognised the
behaviour as
problematic
41 (yes)
7 (no)
Felt responsible for
doing something
34 (yes)
14 (no)
Possessed the correct
skills to take action
3 (yes)
45 (no)
Personal reasons 23 (yes)
25 (no)
Barriers
Not noticing the
behaviour
3 (yes)
13 (no)
Nocondence 8(yes)
8 (no)
Lacking correct skills 3 (yes)
13 (no)
Fear of retaliation 5 (yes)
11 (no)
Not a problem,
no motivation
6 (yes)
10 (no)
Table 14 Details of the sexist banter witnessed

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
63
Witnessed Violence Banter
Yes No
Gender
Man 8 (33.3%) 16 (66.7%)
Woman 38 (27.3%) 101 (72.7%)
X2(1, N=163) = .363
P 0.547
Age
18-34 13 (21.3%) 48 (78.7%)
35-44 18 (39.1%) 28 (60.9%)
45+ 15 (26.8%) 41 (73.2%)
X2(2, N=163) = 4.197
P 0.123
Lockdown Status
Mainly at home 37 (29.4%) 89 (70.6%)
Mainly away from
home 9 (24.3%) 28 (75.7%)
X2(1, N=163) = .359
P 0.549
Table 15 Chi-squared results for participant demographics and witnessing sexual and domestic violence banter
D: Data Tables for Sexual and Domestic Violence Banter

A mixed methods study into bystander experiences of domestic violence and abuse during the COVID-19 pandemic
64
Frequency
Gender of the person
who shared this
Man 35
Woman 5
Genderuid/
non-binary 4
Unsure 1
Since the pandemic
began, have you
noticed this more,
less or about the
same?
More 29
Less 1
About the same 16
Took action?
Yes 33
No 12
Action Taken
Reported the post or
shared disapproval
25 (yes)
20 (no)
Shared concerns
with others
10 (yes)
35 (no)
Looked for more
information
8 (yes)
37 (no)
Took an alternative
form of action
4 (yes)
41 (no)
Motivation for
taking action
Recognised the
behaviour as
problematic
31 (yes)
3 (no)
Felt responsible for
doing something
26 (yes)
8 (no)
Possessed the correct
skills to take action
11 (yes)
23 (no)
Personal reasons 14 (yes)
20 (no)
Barriers
Not noticing the
behaviour
1 (yes)
5 (no)
Nocondence 1 (yes)
5 (no)
Lacking correct skills 3 (yes)
3 (no)
Fear of retaliation 1 (yes)
5 (no)
Table 16 Details of the sexual and domestic violence banter witnessed