Child abduction: understanding police re c o rd e d
Geoff Newiss and Lauren Fairbrother
The offence of child abduction is part of the Violence Against the Person category of police
re c o rded crime. Whilst the offence forms only a small pro p o rtion of all Violence Against the
Person offences (just 0.1% in 2002/03), the total number of offences re c o rded by the police
i n c reased by 45% in 2002/03 from the previous year, to 846 offences (Simmons and Dodd,
2003). The main t y p e s of offence constituting the total number of re c o rded child abductions
a c ross
may have increased so dramatically from the previous year are also discussed.
The views expressed in these findings are those of the authors, not
necessarily those of the Home Office (nor do they reflect Government policy)
l Just over half of all police recorded child abductions were attempted abductions.
l More than half (56%) of all police recorded child abductions involved an offender not known
to the victim (i.e. a stranger).
l Attempted child abductions by strangers were the largest single type of child abductions
(47%). In most offences there was minimal contact between the victim and the offender. The
number of these offences appears to have increased dramatically since 2001/02 – changes
to the manner in which the police record crime are likely to have contributed to this increase.
l Child abductions by strangers that were actually successful, rather than attempts, accounted
for 9% of all offences recorded (a total of 68 victims in 2002/03).
l 23% of child abductions were committed by a parent of the child. In some cases the child
was not taken outside the
should not have been recorded or should have been recorded as ‘no-crime’ at a later stage.
l At least 6% of child abductions (probably considerably more) were sexually motivated.
l 22% of child abductions were classified as ‘other’. In these offences some relationship existed
between the victim and the offender prior to the abduction, including ‘grooming’ style
relationships, familial relationships, friends, abductions for revenge and abductions by boyfriends.
The Research, Development and
Statistics Directorate exists to
improve policy making, decision
taking and practice in support
of the Home Office purpose and
aims, to provide the public and
Parliament with inform a t i o n
necessary for informed debate
and to publish information for
Findings are produced by the
Research, Development and
Statistics Directorate. For further
Communication Development Unit
50 Queen Anne’s Gate,
Tel: 020 7273 2084
Fax: 020 7222 0211
p u b l i c a t i o n s . rd s @ h o m e o ff i c e . g s i . g o v. u k
© Crown copyright 2004
Printed by: TABS
For the purposes of this re s e a rch, offences were
g rouped into four main types:
• parental child abductions – abductions of a
child by a natural parent or guardian
• stranger attempted child abductions –
abductions by someone not known to the
child which did not result in the successful
abduction of the child
• stranger successful child abductions –
abductions by someone not known to the
child that succeeded
• other – successful and attempted abductions
of a child by someone with some pre v i o u s
connection to the child.
Data were collected from all police forces in
Forces were asked to send details of the victims and the
nature of the offence committed. Some offences had since
been re c o rded as ‘no-crime’. In other cases, insuff i c i e n t
detail was available on the victim and/or the offence. The
total number of offences used in this analysis was 798.
The description of each type of child abduction is based
on the number of o ff e n c e s re c o rded by the police (each
victim should be re c o rded as a separate offence). Some
o ffences may be part of the same c a s e, i.e. there was
m o re than one victim. Details are provided at the
beginning of each section on the number of off e n c e s
(victims) and the respective number of cases.
Parental child abductions
23% of the child abductions re c o rded in 2002/03 were
abductions by a parent (141 cases involving 180 childre n ) .
• The average age of the victims was six years – the
lowest for all four types of child abduction (compared
with nine years across all types, see Table 1).
• Slightly more victims (56%) were male (Figure 2).
• Victims of parental abduction were much more likely
to come from ethnic minority groups compared with
other types of child abduction, with just under half of
the victims from non-white ethnic groups (Figure 3).
This reflects previous re s e a rch which indicates that
p a rental abduction of children is closely associated with
p a rtners from diff e rent nationalities, races or cultural
backgrounds disputing the custody of their children (see,
for example, Hegar and Greif, 1994; Plass et al., 1997).
H o w e v e r, the issue is complex and multifaceted,
incorporating problems of domestic violence and abuse as
well as concerns with cultural identity and ways of living
(Plass et al. 1997; Weiner, 2000).
Nearly 90% of the parental child abductions were
successful rather than attempted abductions of the victim.
The offence of child abduction is defined under the
terms of the Child Abduction Act 1984, which allows
for two types of abduction:
Abduction of a child by parent
‘a person connected with the child under the age of
sixteen ... takes or sends the child out of the United
Kingdom without the appropriate consent.’
(Child Abduction Act 1984, Sec 1)
Person connected with the child includes a parent, the
f a t h e r, a guardian, or person with either a re s i d e n c e
order or custody over the child.
Abduction of child by other persons
‘a person other than [the child’s mother, father or
others covered in sec 1], without lawful authority or
reasonable excuse ... takes or detains a child under
the age of sixteen ...’
(Child Abduction Act 1984, Sec 2)
The Principal Crime Rule (Home Office Counting Rules)
applies to offences of child abduction in cases where a
m o re serious crime is committed as part of the same
o ffence. There f o re, if a child is abducted and then
m u rd e red, only one offence of homicide should be
re c o rd e d . Abductions of children for sexual purposes
a re covered in a separate offence categor y of
‘abduction’, although this applies only to female
victims, and is again subject to the Principal Crime Rule
(for example, in cases where an abducted victim is
raped an offence of rape should be recorded).
Figure 1 shows the proportion that each of the four types
accounted for out of the total number of off e n c e s
recorded in 2002/03.
F i g u re 1 Type of police re c o rded off e n c e s
of child abduction (2002/03)
Note: Percentages do not sum to 100 because of rounding.
Crime recording: parental abductions
The Child Abduction Act 1984 provides that the
o ffence of abduction of a child by a parent is
committed only if the child is taken out of the
period exceeding one month (although there are a
number of clauses and caveats which qualify this – see,
for example, Carr and Turner, 2002).
It was not possible to say with complete certainty whether
each offence of parental child abduction collected in this
re s e a rch did meet the legal criteria to be re c o rded as
such. However, it is estimated that at least one-third of
p a rental child abduction offences should not actually
have been re c o rded by the police, or should have been
re c o rded as ‘no-crime’ at a later stage, according to the
legal definition. In some cases, it may have been diff i c u l t
to tell at the time of re c o rding whether the child had
actually been taken out of the
H o w e v e r, in less than 40% of the offences was it clear that
the victim had been taken outside the
30% of the offences it was clear the victim had n o t left the
c o u n t ry). In six offences, the abductor was prevented fro m
removing their child from the country by police at a port .
In those offences where the victim was known to have been
taken outside the
non-white ethnic gro u p s .
Stranger attempted child abductions
In 2002/03 there were 361 cases involving a stranger
attempting to abduct a child or children, resulting in 377
v i c t i m s / o ffences (47% of all child abductions). (These
Table 1 Age of victims by type of child
a b d u c t i o n
Type of Av e r a g e S t a n d a rd N o . * *
a b d u c t i o n a g e * d e v i a t i o n
P a re n t a l 6 4 . 3 1 6 6
Stranger attempted 1 0 3 . 9 3 6 4
Stranger successful 1 0 4 . 5 6 7
O t h e r 1 2 4 . 4 1 6 0
All cases 9 4 . 7 7 5 7
Note: * The average age has been rounded to the neare s t
whole year. ** In 41 cases the age of the victim was not
p ro v i d e d .
F i g u re 2 Sex of victims by type of child
F i g u re 3 Ethnicity of victims by type of
Note: In 46 cases the sex of the victim was not pro v i d e d ,
therefore the total number of cases in Figure 2 is 752.
Note: In 164 cases the ethnicity of the victim was not pro v i d e d ,
t h e re f o re the total number of cases in Figure 3 is 634.
f i g u res include one case of a conspiracy to commit child
abduction where the offenders’ plans to abduct a child
w e re stopped before even an attempt could be made).
• The average age of the victims was ten years, four
years higher than the six year average of victims of
p a rental abduction (Table 1).
• 61% of the victims were female (Figure 2).
• Over 90% of the victims were white (Figure 3).
It was very difficult to establish a motive for these offences
because in many cases the victim had no or only minimal
contact with the offender (in 96% of the stranger
attempted child abductions the motive had to be coded as
‘not known’). For only 15 offences (4%) was there
sufficient information to be certain of the motive – all were
sexually motivated. However, it seems reasonable to
assume that many of the stranger attempted child
abductions where the motive was not known would have
actually been sexually motivated offences. The most
common scenario in these cases was a male off e n d e r
attempting to entice a child into a car or physically
attempting to drag a victim from a public place.
14 offences of stranger attempted child abduction
involved a victim under the age of two years. In each case
the offender attempted to remove the child from the direct
c o n t rol of its parent. In three cases, the off e n d e r
impersonated a member of the social services or health
profession. In the remaining 11 cases the offender tried to
physically remove the victim.
Stranger successful child abductions
In 2002/03 there were 59 cases involving a stranger
successfully abducting a child or children, resulting in 68
victims/offences (9% of all child abductions recorded).
• The average age of these victims was ten years, the
same as the victims of stranger attempted abductions
• 54% of the victims were female (Figure 2).
• 74% of victims (n=39) were white, 17% (n=9) were
black, 8% (n=4) were Asian, and one victim’s ethnic
appearance was coded as ‘other’ (Figure 3).
None of these offences involved the victim being taken
overseas. In all offences where information was available
(63% of the stranger successful offences) the abducted
child was recovered within 24 hours of being taken. There
was no information as to when the remaining 25 victims
were recovered. In most offences, insufficient details were
available to code the motive for the offence re l i a b l y.
However, in 12 offences (19%) it was clear the motive was
sexual. Two victims were subjected to a serious sexual
assault; in the remaining ten offences it was not clear
( f rom the information collected) how serious the sexual
component of the offence had been.
Other child abductions
155 cases of child abduction, resulting in 173
v i c t i m s / o ffences, were classified as ‘other’ (22% of all child
abduction offences re c o rded by the police in 2002/03). In
these cases, some form of relationship existed between the
victim and the off e n d e r.
• The average age of victims was 12 years, three years
older than the average age across all types (Table 1).
• 65% of the victims were female (Figure 2).
• 83% of victims were white, 10% were black and 4%
Asian (Figure 3).
T h e re were several main sub-categories of ‘other’ off e n c e s :
• ‘Grooming’ and exploitative relationships
77 (45%) ‘other’ child abduction offences could be
grouped under the general heading of ‘grooming’
style offences or other exploitative relationships that
adult offenders formed with child victims. This
included offenders offering refuge to young people
who had repeatedly gone missing and enticing them
into using alcohol, drugs and possibly becoming
involved in prostitution. Four of these off e n c e s
involved taking a child overseas. 17% (13 offences)
could with some certainty be classified as sexually
motivated. It seems certain that other offences were of
a sexual nature but insufficient information was
obtained to code them as such reliably. Four offences
involved contact initiated over the Internet – three
were of an overtly sexual nature.
• Abductions by other family members, partners or
Abductions by family members other than the child’s
parents (including friends and partners or ex-partners
of one parent) accounted for a further 32 offences
(18% of the ‘other’ category). Seven off e n c e s
occurred as a result of disputes over the custody of the
victim. Two offences involved taking a child overseas.
• Abductions for revenge/altercation
8% of other offences (n=14) were the result of
ongoing altercations between the victim and off e n d e r
or for the purpose of revenge. Three offences could
be described as ‘punishment beatings’ for off e n c e s
committed by the victim. In nine offences the victim
s u ff e red physical injuries as a result of being
assaulted and in a further three offences the victim
was threatened (in two cases with a weapon).
• Abductions by boyfriends
Nine offences (5% of other offences) involved the
victim going missing with their known boyfriend. All
victims were females aged between 13 and 15.
(These were not classified as sexually motivated child
abductions for the purposes of this study.)
Crime recording: sexual offences
Across all types of child abduction offences recorded in
2002/03, a total of 45 offences (6%) could reliably be
classified as sexually motivated. 32 of these off e n c e s
involved female victims. This is likely to be a
considerable underestimate of the actual number of
sexually motivated offences within the total number of
child abduction offences recorded by police (in 96% of
the stranger attempted child abductions it was not
possible to classify the motive).
Whilst it may have been difficult for the police to
establish the exact motive at the time the offence was
re p o rted, it is possible that at least some of these
o ff ences would hav e bee n classi fi ed more
a p p ropriatel y as of f ences of abduc tion (s ee
Definitions box) rather than chi ld abduction.
H o w e v e r, this is an area where crime re c o rding is
like ly to c hange over the next y ear with the
i n t roduction of a broader 'pre p a r a t o r y' sexual
o ffence. Section 62 of the Sexual Offences Act 2003,
which received Royal Assent in November 2003 and
should come into force in May 2004, provides for
the offence of "committing an offence with the intent
to commit a sexual offence" and would include
a b d u c t i o n .
Why have re c o rded offences of child
Police recorded offences of child abduction increased by
45% from 2001/02 to 2002/03 (Simmons and Dodd,
2003). In an attempt to provide some explanation for this
i n c rease, data were collected on child abductions
re c o rded in the previous year (2001/02) from the five
f o rces which re c o rded the largest numerical increases in
offences. Together, these forces accounted for 57% of the
total increase across all forces from 2001/02 to
2002/03, and recorded just over one-third of all offences
of child abduction in
Table 2 shows how the type of child abductions recorded
changed in these five forces.
Whilst the actual number of all types of child abduction
i n c reased across the five forces, the largest increase was in
stranger attempted abductions. The actual number of
stranger attempted child abductions re c o rded across the
five forces increased by nearly 200% (from 43 offences in
2001/02 to 128 offences in 2002/03). As a pro p o rtion of
child abduction offences re c o rded in the five forc e s ,
stranger attempted child abductions increased from less
than one-third in 2001/02 to nearly half in 2002/03. This
matches the pro p o rtion of stranger attempted abductions
re c o rded by all forces in 2002/03 (see Table 2).
The number of parental abductions increased by nearly
t w o - t h i rds across the five forces from 56 offences in
2001/02 to 92 offences in 2002/03. However, as a
p ro p o rtion of all child abduction offences re c o rded in the
five forces in 2002/03 parental abductions decre a s e d
f rom 39% to 33% (the rise in the actual number of this type
of abduction was offset by the much larger rise in stranger
attempted child abductions). This gives a clear indication
that stranger attempted abductions are the primary re a s o n
for the large increase in all child abductions re c o rded in
2002/03, followed by the rise in parental abductions.
The large increase in re c o rded stranger attempted child
abductions may, at least in part, be attributable to the
general move to a more prima facie approach to the
re c o rding of crime (based on the victim’s perception of a
crime occurring). This has had an impact in several are a s
of crime re c o rding, not least violent crime. Simmons, Legg
and Hosking (2003) describe the impact of the National
Crime Recording Standard (NCRS) – a standard adopted
did implement the same practices earlier. It is feasible that
re p o rts of attempted child abductions where no contact
was made with the victim may have gone unre c o rded as
crimes prior to the introduction of the NCRS. The latest
f i g u res there f o re may provide a truer reflection of the
extent of child abduction in
The move to a more victim-focused approach to crime
re c o rding may have led to some offences of pare n t a l
abduction being recorded which fell outside the scope of
the legal definition of this offence. This might be coupled
with a very real increase in the number of cases of
p a rental abduction, although at this stage there is
insufficient data to provide firm evidence of such a trend.
There is anecdotal evidence that officers are increasingly
better trained to recognise the potential seriousness of
p a rental abductions and there f o re not to consider them
solely as ‘domestic’ issues which lie outside the scope of
police involvement (Hillier, 2000).
Detection and identification of the suspect
The detection rate for child abduction offences fell fro m
50% in 2001/02 to 37% in 2002/03 (compared to an
average for all Violence Against the Person offences of
58% in 2001/02 and 54% in 2002/03). This re s e a rc h
did not examine the actual disposal code for each offence
re c o rded in 2002/03 (to do so would have re q u i re d
many forces to examine more than one recording system).
H o w e v e r, data were collected on whether a suspect had
been identified for the offence:
• in all parental abductions and 90% of ‘other’
abductions a suspect was identified
• in 54% of stranger successful abductions a suspect
had been identified
• in only 13% of stranger attempted abductions a
suspect been identified.
Table 2 Type of child abduction offences re c o rded in five forces, 2001/02 to 2002/03
Type of abduction Proportion of cases by type of abduction
2 0 0 1 / 0 2 2 0 0 2 / 0 3 2 0 0 2 / 0 3
Five forc e s Five forc e s All forc e s
% N o . % N o . % N o .
P a re n t a l 3 9 5 6 3 3 9 2 2 3 1 8 0
Stranger attempted 3 0 4 3 4 6 1 2 8 4 7 3 7 7
Stranger successful 8 1 2 6 1 6 9 6 8
O t h e r 2 2 3 2 1 6 4 4 2 2 1 7 3
To t a l 1 0 0 1 4 3 1 0 0 2 8 0 1 0 0 7 9 8
Note: Percentages do not add to 100 due to ro u n d i n g .
G e o ff Newiss is a Senior Research Officer and Lauren Fairbrother is a Research Trainee in the Crime and Policing
Group, Home Office Research, Development and Statistics Directorate.
C a rr, A.P. and Tu rn e r, A.J. (eds.) (2002). Stone’s
Justices’ Manual. Volume 1, pp.1964-1968.
Lexis Nexis Butterworths.
Erikson, M. and Friendship, C. (2002). A typology of
child abduction events. Legal and Criminological
Psychology, 7, 2002, pp.115–120.
H e g a r, R. L. and Greif, G. L. (1994). Pare n t a l
abduction of children from interracial and cro s s -
cultural marriages. J o u rnal of Comparative Family
Studies, 25, (1), Spring 1994, pp.135–142.
Hillier, J. (2000). Foreign Affairs. Police Review, 9 June
Plass, P.S., Finkelhor, D. and Hotaling, G.T. (1997).
Risk factors for family abduction: demographic and
family interaction characteristics. J o u rnal of Family
Violence, 12, (3), September 1997, pp.333–348.
Simmons, J. and Dodd T. (eds.) (2003). Crime in
Statistical Bulletin 07/03.
Simmons, J., Legg, C. and Hosking, R. (2003). T h e
National Crime Recording Standard (NCRS): an
analysis of the impact on recorded crime. Home Office
On-Line Report 32/03.
We i n e r, M.H. (2000). International child abduction
and the escape from domestic violence. Fordham Law
Review, LXIX, (2), pp.593–706.
Given that stranger attempted child abductions are likely
to have accounted for a large part of the overall increase
in child abduction offences (owing at least in part to the
effect of the NCRS), this could provide a good indication
of why the detection rate has fallen.
Whilst suspects were identified in all cases of pare n t a l
abduction but just over half of stranger successful
abductions, previous re s e a rch has shown that o ff e n d e r s
c o n v i c t e d of child abduction are mainly sexually
motivated and are not related to the victim (Erikson and
Friendship, 2002). This is likely to reflect a higher
p ropensity of the police and Crown Prosecution Serv i c e
to pursue a prosecution against an offender who abducts
a child not known to him or her, particularly for sexual
purposes. Whilst it was not possible to produce re l i a b l e
data on whether the police pursued a case against a
known off e n d e r, it was clear that of the 2002/03 cases
of parental child abduction, the vast majority did not
result in prosecution. Information provided in one police
f o rce indicated that the police tended to encourage
estranged parents to pursue a formal custody settlement
in the civil courts.
• The Home Office, through the National Crime
Recording Steering Group (NCRSG), should review
the instructions for the recording and reclassification
of child abductions by parents to ensure that only
appropriate offences are recorded by the police.
Guidance should be issued if necessary.
• The Home Office, through the NCRSG, should review
the instructions for the recording and reclassification
of child abductions of a sexual nature, and should
consider issuing guidance to police forces to clarify
the specific circumstances in which an abduction of a
child for sexual purposes should be recorded as a
child abduction (Violence Against the Person
o ff e n c e) or an abduction (sexual offence). This review
could be undertaken as part of the implementation of
the prospective new sex offences legislation.
• The Home Office should monitor the number of child
abductions recorded by the police closely and should
undertake further work on why these offences are
increasing, if this continues to be the case.
• The Home Office should liaise with the Foreign and
Commonwealth Office and the Department for
Constitutional Affairs to monitor patterns and trends
in parental child abductions overseas.
The Child Exploitation and Online Protection Centre.
Child Exploitation and Online Protection Centre
Help Find Madeleine McCann
Can You Help?
British police are calling on members of the public to help further in the search for Madeleine McCann. If you were at the Ocean Club Resort, Praia da Luz or in the surrounding areas in the two weeks leading up to Madeleine’s disappearance on Thursday 3 May then we would like to have copies of any photographs taken during your stay.
In particular we would like any pictures that have people in them who you don’t know as opposed to scenery shots or pictures of just your own family. If you have any such pictures then upload them here.
If you are unsure how to upload your pictures then click on this link where you will find a simple how to guide. You can also go into any high street photographic retailer and ask them to help in uploading your pictures onto this website.
All photographs will be treated in the strictest confidence and will be used for investigative purposes only.
British and Portuguese police are also asking anyone to contact them on the numbers below if they have seen anything suspicious related to the disappearance of Madeleine or if you believe you know where Madeleine is being concealed or hidden. Portuguese police have searched extensively around Praia da Luz and she has not been found. It is possible that she is being hidden or concealed in some way and if you know where then by now you may have realised that it is in everyone’s interest that she is returned to her family.
British police on 0800 096 1233
This website has been delivered by the combined