Katherine M. Brown,1 M.A. and Robert D. Keppel,2 Ph.D.
Child Abduction Murder: An Analysis of the
Effect of Time and Distance Separation Between
Murder Incident Sites on Solvability
ABSTRACT: Empirical studies of child abduction murder investigations are lacking. Accordingly, an empirical analysis of the effect of time and
distance relationships on case solvability in child abduction murders (N 5 735) was conducted. The murders occurred across the United States
from 1968 to 2002. Murder incident components examined were: victim last seen site, initial contact site, murder site, and body recovery site.
Time and distance intervals between component pairings were also examined. Descriptive, bivariate, and multivariate analyses were performed to
determine if information relating to time and distance intervals between components were critical solvability factors. Results show that
information about time and distance increases case solvability. Results also demonstrate that time and distance relationships contribute uniquely to
case solvability in murders of abducted children. Findings also indicate that additional factors such as type of forensic evidence, investigative
resources, or actions by first responders, may be critical to case solvability.
KEYWORDS: forensic science, murder of missing and abducted children, murder, missing and abducted children, solvability, time, distance,
clearance, abduction, kidnapping, homicide
Child abduction is every parent’s worst nightmare. The only
thing more terrifying and traumatic than a child abduction is the
murder of an abducted child. To compound the problem, child
abduction murders are incredibly difficult to solve and deeply
impact law enforcement officials involved in the investigation.
Abductions which result in a child’s death present great investi-
gative and emotional obstacles for law enforcement officers (1).
The rarity of child abduction murders, even among criminal
homicides, and their complex, emotion-laden, high profiles, are
extremely difficult to investigate (2).
The 1979 abduction of 7-year-old Etan Patz, and the 1981
abduction and murder of 6-year-old Adam Walsh terrified parents
and devastated the nation. The public outcry over concern for the
safety of America’s children resulted in the establishment of the
National Center for Missing and Exploited Children (NCMEC) by
the United States Congress. Unfortunately, as a result of these
events, inflated and unsubstantiated numbers of missing children
were widely reported to be over 2 million per year during the mid
to late 1980s. The numbers of children who were abducted and
then murdered were erroneously reported to be as high as 5000 per
year (1).
Studies show that children are murdered in 40% of stereotypical
abductions and not recovered in an additional 4% of abductions
(3). Fortunately, this occurrence is rare. Approximately 40–150
incidents of child abduction murder occur across the nation each
year (2). This figure is less than one half of 1% of all murders
committed in this country every year (2). Previous studies have
shown that in c. 74–76% of abducted child cases, the child was
killed in less than 3 h (2,4). Studies also showed that while 22% of
the children were still alive at the time they were reported missing,
42% had already been killed before they were reported missing
(2). Obviously, time is critical in the report of any missing child to
authorities.
While changes in our public policies and laws have made
important progress toward protecting our children and keeping
families in tact, a vital part of the murder of an abducted child
investigation, solvability, has not been adequately addressed by
researchers. The dearth of research in this area may be due to
limitations in access to sensitive law enforcement data, or lack of
expertise in the area. However, because the murder of an abducted
child impacts our society in such an overwhelming manner, the
absence of literature in this area is disturbing.
Challenges to Investigation
The rarity of child abduction murder creates special challenges
for homicide investigation (2). Agencies with limited resources
can become quickly overwhelmed when a high profile case occurs
(5,6). In addition, the high profile nature of previous cases has
created commonly held false assumptions and beliefs about the
murder of abducted children (2). Child abduction murder inves-
tigations involve multiple agencies, investigators, crime scene
technicians, forensic experts, and prosecutors with skill levels by
agency varying. This diversity can lead to communication pro-
blems and the failure to recognize that some cases may be linked
(7). Failure to recognize linked cases is important because
offenders transported their victims and disposed of their bodies
in a different jurisdiction in 36% of cases (2). Unfortunately, very
little has been written about the effectiveness of child abduction
investigations specifically and law enforcement investigations
generally (2,7–10).
Solvability
Solvability is the ability of investigators to identify the perpe-
trator of a crime. The solvability of a case is not dependant on case
1
College of Criminal Justice, Sam Houston State University, Huntsville,
TX 77341-2296.
2
Department of Criminal Justice, Seattle University, Seattle, WA 98122-
1090.
Received 15 April 2006; and in revised form 19 July 2006; accepted
13 Aug. 2006; published 10 Dec. 2006.
137Copyright r 2006 by American Academy of Forensic Sciences
J Forensic Sci, January 2007, Vol. 52, No. 1
doi:10.1111/j.1556-4029.2006.00328.x
Available online at:
www.blackwell-synergy.comclearance. Research has shown that typically a suspect is in
custody within 24 h in 66% of murder cases. Further research
indicates that if a murder is not solved within 48 h, the solvability
of the murder decreases dramatically (10,11). It is clear that time
and distance are critical factors in case solvability, yet little
empirical research has been conducted in this area (12–16).
Effect of Time and Distance on Case Solvability
The FBI Behavioral Science Unit has addressed time and
distance as important factors in profiling violent offenders. The
research emphasizes the importance of the time it takes an
offender to murder and dispose of the victim in relation to the
location where the murder occurred. The research also indicates
the importance of the location if the abduction point is different
from the body recovery site (17). However, location as a solva-
bility factor in child abduction murders and serial murder cases
has been explored by few researchers (18).
A 1990 study by the U.S. Office of Juvenile Justice and
Delinquency Prevention examined the time of detention and the
distance of transportation after abduction. The research indicates
that 2% of abduction cases ended in murder where children were
transported over 20 feet or detained for more than an hour.
Unfortunately, this study did not address the relationship of time
and distance as solvability factors in those cases (19).
Components of a Murder Incident
Perhaps the most comprehensive study on the relationship of
time and distance as solvability factors was conducted by Keppel
and Weis (20). The study examined the relationship between
solvability and time and distance separation between the murder
incident sites in homicide investigations. Keppel and Weis con-
ceptualize the crime of murder as an incident (20). Each murder
incident may include multiple sites or locations of contact
between the offender, or a witness, and the victim. The investiga-
tion into a murder emphasizes the search for clues or information
about the major investigative sites of a murder incident. The
presence of information that establishes the existence of each site,
coupled with when and where each site is located within the
incident, and the manner in which their relationships affect each
other, greatly influences the solution of murder investigations
(20,21).
The follow-up investigation of a murder involves the gathering
of information about the various components that are locations of
victim–offender contact. The types of information crucial to the
investigation in order of their usual occurrence within the murder
event are:
1. where and when the victim was last seen,
2. where and when the offender initially contacted the victim,
3. where the murder took place, and
4. where and when the body was recovered (20).
Each of these components occurs in a murder and each component
is important in a murder investigation. Information about the
location and the time of each site within the sequence of the
murder incident has an impact on case solvability. In most murder
cases, the events occur simultaneously. Research suggests in
general murder; events are located in the same place and not
separated by intervals of distance or spans of time (2,20,21).
Separation of Components by Time and Distance
The sites within most incidents of murder may become sepa-
rated by time and distance. The separation may occur in two ways.
First, the offender may consciously separate time and distance.
The killer may believe that the separation of murder components
will delay the discovery of various sites contributing to the
destruction of evidence. For instance, the killer may intentionally
abduct a child and use a vehicle to transport the child to a remote
location to murder the child and yet another location to dispose of
the body. As a result, the separation of components by time and
distance may inhibit the investigation by creating communication
and cooperation problems among police agencies because the
locations of all sites are not within the authority of one police
agency.
Second, the offender may unintentionally separate the location
of components by time and distance. For example, the killer may
transport the child with a car because he was unable to lure the
child inside by the ruse he planned. Another example would be the
transport of the child to another location because the killer would
rather not have someone see the sexual assault of the child. These
actions do not show a deliberate effort by the killer to delay
discovery of the body or conceal evidence, but are simply
strategies to avoid detection.
Finally, the time and location of murder incident components
may be separated by pure chance (2,20,21). This type of delay
would be a situation or event, not in control of the killer. For
example, perhaps the killer chose a body disposal site off of a road
that was subsequently blocked from access for road repairs. This
type of event might delay the discovery of the body, but happened
by chance, rather than design of the killer.
Keppel and Weis determined that in general murders when any
information on the dates and locations of the murder incident sites
is known, the probability of case solution increases (20). The
study also found a strong positive correlation between knowing
the dates of occurrences for the murder incident locations and the
ability to identify a perpetrator (20). Case solvability increases as
the time between pairs of murder incident sites decreases (20). In
addition, the more investigators know about the distances between
the pairs of the murder incident sites, the more case solvability
increases (20). When the distance between the locations is less
than 199 ft, solvability increases (20). Finally, the study indicates
that when the time and distances decrease among pairs of murder
incident sites, the solvability of a case increases (20,21).
Keppel and Weis empirically supported what experienced
murder investigators know by instinct (20). The more information
detectives know about a case; the higher the case solvability will
be. However, their study indicates that it is not just any informa-
tion that will enhance solvability—some information is more
valuable and useful than other information in murder investiga-
tions (20). There is a clear need for more multivariate analyses to
study the role of time and distance as solvability factors in murder
investigations. As comprehensive as their study was, Keppel and
Weis did not address the factors of time and distance and their
relationship to solvability in murder investigations of abducted
children (20).
Separation of Components by Time and Distance in Child
Abduction Murders
A subsequent study of murders of abducted children by the
Washington State Attorney General’s office did identify a unique
pattern of distance relationships in child abduction murders (2).
138 JOURNAL OF FORENSIC SCIENCES
Often, the initial contact site is located very close to the victim’s
last known location. Importantly and conversely, the distance
between the initial contact site and the murder site increases to
distances greater than 1/4 mile in murders of abducted children.
The distance from the murder site to the body recovery site
decreases to less than or equal to 199 feet in the vast majority
of cases (2).
Time and distance have been examined as part of solvability
research for murders in general. However, before this study, the
relationship to case solvability has not been examined in murder
investigations of abducted children. This research will be used to
improve the efficiency and effectiveness of the investigation
processes of those murders. The research results from this study
are particularly salient to homicide that detectives confronted with
an unsolved murder case involving child abduction. In addition,
results from this study will help police investigators to efficiently
identify strategies and implement tactics which will lead to the
capture of child abduction killers and the solution of cases.
Methods
The objective of the collaborative research project was to
examine solvability factors in murder investigations of abducted
children. The data were analyzed to determine if time and distance
relationships between murder incident component pairings would
contribute to case solvability in murder investigations of abducted
children. The data used in this study were collected through a
cooperative agreement between the Washington State Attorney
General’s Office and the United States Department of Justice
Office of Juvenile Justice and Delinquency Prevention. This
dataset will be referred to as the child abduction murder (CAM)
dataset.
The CAM data is neither a random sample of CAM cases nor a
sample of all CAM occurring in the United States. Data were
collected based on voluntary reporting of cases from each law
enforcement jurisdiction contacted. Initially, data were collected
by interviews with the detectives and the review of investigative
case files from 1025 cases. Responses were received from both
large and small agencies and departments. The agencies were
representative of all regions of the country and 44 states.
Of the original 1025 cases, 621 cases were found to meet the
criteria established for inclusion in the original data collection
collaboration (2,4). Subsequent cases have been collected for
inclusion in the original dataset resulting in 833 child abduction
murder cases from murders committed from 1968 to 2002 (4).
Case Criteria
The cases of murder in the original CAM data collection were
chosen for inclusion based on the following criteria:
1. The victim was younger than 18 years old (except as described
in #3 below), whose body had been recovered, or if the body
had not been recovered, the killer was identified, tried, and
convicted; and
2. The police agency receiving the initial contact about the case,
whether as a missing, abducted, runaway, or dead body case,
acted on the premise that abduction was a possibility;
3. The case was part of a series in which at least one victim in the
series met the above-stated criteria (2).
Additionally, cases were included in this dataset that were not
considered closed in the traditional sense. If the reporting agency
believed that abduction was a possibility and began investigating
the case as a child abduction case, it was included in the data.
The cases of murder of abducted children to be examined in this
research project were selected from the CAM dataset in which the
victim was 17 years old or younger, whose body had been
recovered, or if the body had not been recovered, the killer was
identified, tried, and convicted. On examination, 735 cases in the
CAM Dataset met the criteria and were used in this research
project.
Definitions
In order to select appropriate variables for analysis in our study,
certain terminologies were defined: abduction, components of the
murder incident, time and distance intervals, and solved cases.
Once the terminology was defined and the proper parameters set,
then appropriate variables were selected for statistical analysis.
Abduction
Defining the terms used in this research project was critical. For
instance, the word ‘‘abduction’’ can be interpreted from several
different perspectives. For purposes of this research, abduction in
the original data collection was defined as
1. The victim was kidnapped,
2. The victim was detained and his or her freedom of movement
was restricted,
3. A victim of domestic violence was reported by the family
(or someone else) as a missing child, and
4. The police were initially of the opinion that the victim was
taken or held against his or her will, whether or not that turned
out to be the case in the end (2).
Components of the Murder Incident
1. The victim last seen (VLS) site was defined as the location
where and time when the victim was last seen. The VLS was
determined from eyewitness information and records indicat-
ing when and where the victim was last seen alive.
2. The initial contact (IC) site was defined as the place where and
time when the killer initially contacted the victim. The IC was
established from evidence indicating that the killer first met the
victim at a certain time and at a specific location during the
course of the murder incident.
3. The murder site (MS) was defined as the place where and time
when the victim sustained the death-producing injuries. The
MS was established from evidence, confession of the offender,
or other information provided by detectives.
4. The body recovery (BR) site was defined as the location where
and time when police, medics, or witnesses found the victim,
dead or alive, prior to transportation to a medical facility or
morgue (20,21).
Time and Distance Intervals
Time spans between the murder incident sites were examined
by calculating the duration of time from one murder component to
each of the other components. The length of the separations was
measured in hours and minutes. There are six possible pairs of
components for which a time span was calculated:
1. VLS to IC,
2. VLS to MS,
3. VLS to BR,
BROWN AND KEPPEL . CHILD ABDUCTION MURDERS: TIME AND DISTANCE SEPARATION EFFECT ON CASE SOLVABILITY 139
4. IC to MS,
5. IC to BR,
6. MS to BR.
Spans of Time Between Locations
The actual time span between murder incident components was
examined. The metric time spans between the VLS site to IC site,
VLS site to MS, and IC site to MS were then converted into one of
the following time interval categories:
1. 0–29:59 min,
2. 30–59:59 min,
3. 1–4:59:59 h,
4. 5–24 h,
5. 424 h.
Because of differing frequency distributions, the VLS site to BR
site, IC site to BR site, and MS to BR site pairings’ metric time
spans were converted into one of the following categories:
1. 0–7:59 h,
2. 8–15:59 h,
3. 16–23:59 h,
4. 24–47:59 h,
5. 48–71:59 h,
6. 72–167 h,
7. 7–14 days,
8. 15–30 days,
9. 430 days.
These categories were based on the natural breaks in the frequency
distribution of the CAM data. Once this comparison was made, the
time spans were collapsed into two intervals:
1.
24 h,
2. 424 h.
Distance Between Locations
In the initial data collection, the distance between each pair of
murder components was measured in feet or miles for each pair of
components. Then the actual distance was placed into one of the
following categories before inclusion in the CAM dataset:
1. 0–199 ft,
2. 200 ft–o1/4 mile,
3. 1/4–o1 1
2 mile,
4. 1 1
2 –o12 miles,
5.
12 miles.
Once this comparison was made, the distance intervals were
collapsed into two intervals for further statistical analysis:
1.
199 ft,
2. 4199 ft.
The first category was based on the collective experience of
several homicide detectives originally consulted by Dr. Robert
Keppel (21).
Solvability
In the CAM data set, solvability was defined two ways: ‘‘Has
the offender been arrested, or does probable cause exist for an
arrest?’’ and ‘‘Has the investigation resulted in a conviction?’’ For
this research, solvability was based on investigations resulting in
an arrest. Cases which resulted in a ‘‘Yes’’ to the question of ‘‘Has
the offender been arrested, or does probable cause exist for an
arrest?’’ at the time of coding were considered solved for this
research project and cases which answered ‘‘No’’ to that question
were considered unsolved. Cases coded with ‘‘Unknown’’ as an
answer to the question were considered to have missing data and
were not included in the analysis.
Results
Because information known about each murder incident com-
ponent is critical to murder investigations of abducted children,
the components were examined by knowledge of time and
distance and separation between components. The variables used
for this analysis captured the date of occurrence (exact or
approximate), and the type of location and/or the address of that
location. The following sections outline what information about
the murder incident component was known on time, location, or
both in the investigations of child abduction murders contained in
this dataset.
As shown in Table 1, the percentage of unsolved murder cases
in the CAM dataset was 27.4%. It is important to note that the
overall percentage of child abduction murder case solvability for
victims, 17 years old and under, is 72.6%. This is slightly lower
than the solvability for murder victims in general which has been
previously shown to be 77% (21). A reason for this difference
could be that adults which were part of a series that involved
children were excluded.
Importance of Information on Time and Place
Table 2 outlines what information about the murder incident
component was known on time, location, or both, in the investiga-
tions of CAM. The VLS site is the location about which time is
most often known (98.9%). The offender informed police where
the MS was or confirmed its location in their statement to police in
55.9% of cases. However, the actual MS is the location least
identified by time or place.
This order is unique to child abduction murder investigations.
In a general murder, the location of the BR site is the location
most often known, followed by the MS, VLS site, and the IC site
(21). The data indicates that in child abduction murders the order
switches to the VLS site as the location about which information
is most likely known. This is likely due to the nature of the victim.
Children are missed more quickly than adult victims in some
cases, especially small children who are usually in the presence of
a caregiver. It is also not surprising that in these types of
investigations that the percentage of knowledge about murder
incident locations is higher than in general murders because of the
high-profile nature of these cases.
An examination of the murder incident components about
which both time and place were known produced a decrease in
the known percentages. This is not surprising given that much of
TABLE 1—Child abduction murder (CAM) case solvability.
All Cases in CAM Dataset Victims
17 Years Old
n % n %
Solved 589 74.1 527 72.6
Unsolved 206 25.9 199 27.4
Unknown 38 4.6 9 1.2
N 833 735
140 JOURNAL OF FORENSIC SCIENCES
this information is gathered from witness statements or statements
by the killer. The murder incident component about which both
time and location is most often known in child abduction murders
is the BR site (98%), followed in order by the VLS site (77.7%),
IC site (72.2%), and the MS (62.6%) as indicated in Table 2.
Impact of Time on Case Solvability
It was predicted that case solvability will be greater when the
time is known for each location than when the time is unknown in
a murder investigation of an abducted child—When the time of
the murder is known, solvability is 80.2%. When the time of the
murder is unknown, solvability decreases to a mere 31.6%. The
time of the initial contact also exhibits strong solvability (79.7%).
When the time of the initial contact between the victim and
offender is unknown, solvability decreases to 23.9%. Table 3
illustrates case solvability when the time of occurrence for each
location is known.
The knowledge of the time when the victim was last seen and
when the body was recovered is shown to have no relationship to
case solvability (see Table 3). However, the knowledge of the
time of the initial contact between the victim and offender and the
murder is shown to have a significant relationship to case
solvability. This finding makes the significance of locating in-
formation about when the initial contact between the victim and
offender and when the murder occurs of paramount importance to
investigators. The difference in solvability when time is known for
each murder incident component is more striking when shown in
graphic form (see Fig. 1).
VLS
It is important to note that the time span between when the
victim was last seen and the initial contact with the killer is less
than 30 min in over three-fourths of the child abduction murder
investigations. The time span is illustrated in Table 4. The data
also indicate a very short time frame between when the victim was
last seen and the murder. The abducted child is killed in less than
5 h in 82.9% of cases as illustrated in Table 4. An abducted child is
murdered in less than 30 min after he or she was last seen in 30.8%
of cases. The bodies of abducted children are recovered within
24 h of when the victim was last seen in 37.6% of cases, within
48 h in 49.4% of cases and within a week in 68.5% of cases (see
Table 4).
IC
The data also indicated a short time frame between when the
initial contact between the victim and the killer occurred and when
the victim was murdered. The abducted child is murdered in less
than 5 h after the initial contact with the offender in 85.1% of
cases as illustrated in Table 4. The victim is murdered in less than
30 min after the initial contact with the offender in 38.1% of cases.
BR
The victim’s bodies are recovered within 24 h of when the
offender and victim had their initial contact in 39% of cases,
within 48 h in 52.6% of cases and within a week in 73% of cases.
The time frame between the initial contact between the victim and
offender and when the victim’s body is recovered is illustrated in
Table 4. The bodies of abducted children are recovered within
24 h of the murder in 46.4% of cases, within 48 h in 57.6% of
cases and within a week in 76.3% of cases.
Time Proximity
It was predicted that (a) when the time between a given pair of
locations is less than 24 h, such a relatively close proximity in time
TABLE 2—Knowledge of time and place of components.
Murder Incident
Component
Percent
Time or
Place Known
Percent
Time
Known
Percent
Place
Known
Percent Both
Time and Place
Known
Victim last seen site 99.3 98.9 78.1 77.7
Initial contact site 93.5 87.5 78.2 72.2
Murder site 91.3 84.5 69.4 62.6
Body recovery site 99.0 98.6 98.4 98.0
TABLE 3—Relationship between knowledge of time and case solvability.
Murder Incident Component
Time
Known
Percent of
Cases Solved n Tau b p
Victim last seen site Yes 73.0 524 0.083 0.094
No 37.5 3
Initial contact site Yes 79.7 505 0.416 o0.001
No 23.9 22
Murder site Yes 80.2 491 0.397 o0.001
No 31.6 36
Body recovery site Yes 72.3 518
0.046 0.111
No 90.0 9
FIG. 1—Graphic representation of solvability when time is known.
TABLE 4—Spans of time between components.
Time Interval
Percent of Cases
VLS to IC VLS to MS IC to MS
0–29:59 min 76.3 30.8 38.1
30–59:59 min 7.9 16.5 15.2
1–4:59 h 10.8 35.6 31.8
5–24 h 4.1 13.5 11.5
424 h 0.9 3.6 3.4
VLS to BR IC to BR MS to BR
0–7:59 h 17.9 20.1 26.1
8–15:59 h 11.0 10.0 11.4
16–23:59 h 8.7 8.9 8.9
24–47:59 h 11.8 13.6 11.2
48–71:59 h 6.4 6.1 5.8
72–167 h 12.7 14.3 12.9
7–14 days 8.7 7.2 6.1
15–30 days 6.1 5.4 4.3
430 days 16.7 14.5 13.2
VLS, victim last seen; IC, initial contact; MS, murder site; BR, body
recovery.
BROWN AND KEPPEL . CHILD ABDUCTION MURDERS: TIME AND DISTANCE SEPARATION EFFECT ON CASE SOLVABILITY 141
will contribute to the solvability of a case involving the murder of
an abducted child; the percentage of cases solved will be greater
than when the pair of locations is separated by greater than 24 h.
(b) The time proximity of locations will contribute to the solva-
bility of the case even if the locations are not close in time—
because of insufficient frequencies in the time interval categories,
only the following murder incident components were examined
for this hypothesis:
1. VLS site/BR site,
2. IC site/BR site,
3. MS/BR site.
Kendall’s Tau-b test indicates that part (a) of the research
hypothesis is supported for the MS/BR site pairing. There is a
statistical significance when the time between the murder and
body recovery is less than 24 h solvability decreases. When the
time interval is greater than 24 h, 78% of cases were solved while
22% of cases separated by less than 24 h remained unsolved (see
Table 5).
Part (b) of this research hypothesis theorized that the time
proximity of locations will contribute to the solvability of the case
even if the locations are not close in time. This portion of the
hypothesis is supported for the MS/BR site pairing. When these
locations are separated by more than 24 h case solvability in-
creases to an astounding 91.3%, while solvability of unsolved
cases decreases to a mere 8.7%. This suggests that in cases where
the offender does not dispose of the body within 24 h of the
murder, more incriminating physical evidence may be present
which may increase case solvability in murders of abducted
children. The time span between other murder incident component
pairings shows no significant association with case solvability.
Impact of Distance on Case Solvability
It was predicted that when police investigators know the
distance between any pair of locations, this knowledge will
contribute to the solvability of a case; the percentage of murder
of abducted children cases solved will be greater given this
knowledge than when the distances between pairs of locations
are not known—the relationship between case solvability and
knowledge about the distance between the murder incident com-
ponent pairings is shown to have a positive association in all
pairings except for the VLS site/BR site pairing. Results from
Kendall’s Tau-b test support the prediction that the knowledge of
the distance interval between the murder incident component
pairings increases case solvability in murder investigations of
abducted children (see Table 6).
When investigators know the distance interval between murder
incident components, the strongest indicator of solvability is the
IC site/MS pairing (84.7%). Each of the remaining four pairings
also exhibit high solvability percentages. The MS/BR site pairing
closely follows with a solvability percentage of 82%. The VLS
site/MS also exhibits a strong solvability percentage of 81.3%.
The VLS site/IC site pairing indicates a solvability percentage of
79.7% and the IC site/BR site pairing shows a solvability
percentage of 79.5%.
When investigators do not know the distance interval between
the IC site/MS pairing, solvability decreases to 27%. When this
information is not known for the MS/BR site pairing, solvability
diminishes to 29.8%. The VLS site/MS site also exhibits a
solvability decrease to 28%. The VLS site/IC site pairing indicates
a solvability decrease to 21.3% and the IC site/BR site pairing
shows a solvability decrease to only 22.7%. The difference in sol-
vability when distance is known for each murder incident compo-
nent is more striking when shown in graphic form (See Fig. 2).
Distance Proximity
The distance between the IC site and the VLS site is less than
199 ft in 64.2% of cases. The distance between locations is shown
in Table 7. The distance between the VLS site and the MS is less
than 199 ft in 26.6% of CAM cases. The victim’s body is
recovered less than 199 ft where he or she was last seen in only
15.8% of cases. The victim’s body is recovered overone and one-
half miles away from where the initial contact with the offender
occurred in 53.5% of cases. The distance between the IC site and
the MS is less than 199 ft in 34% of cases. The distance between
the MS and the BR site is less than 199 ft in 69.1% of cases.
TABLE 5—Relationship between time span and case solvability.
Murder Incident
Component Pairing
Time
Span
Percent of
Cases Solved n Tau b p
Victim last seen site to
body recovery site
24 h 70.7 152 0.000 0.999
424 h 70.7 251
Initial contact site to body
recovery site
24 h 78.9 131 0.097 0.055
424 h 86.3 220
Murder site to body
recovery site
24 h 78.0 142 0.185 o0.001
424 h 91.3 188
TABLE 6—Relationship between knowledge of distance and case solvability.
Murder Incident Component
Pairing
Distance
Known
Percent of
Cases
Solved n Tau b p
Victim last seen site to initial
contact site
Yes 79.7 508 0.429 o0.001
No 21.3 19
Victim last seen site to
murder site
Yes 81.3 494 0.441 o0.001
No 28.0 33
Victim last seen site to body
recovery site
Yes 73.0 512 0.053 0.203
No 60.0 15
Initial contact site to murder
site
Yes 84.7 486 0.526 o0.001
No 27.0 41
Initial contact site to body
recovery site
Yes 79.5 507 0.415 o0.001
No 22.7 20
Murder site to body recovery
site
Yes 82.0 488 0.450 o0.001
No 29.8 39
FIG. 2—Graphic representation of solvability when distance is known.
142 JOURNAL OF FORENSIC SCIENCES
The distance is greater than one-fourth mile between the VLS
site and MS (54.2%) and the IC site and the MS (51.8%) in just
over one-half of cases. A majority of the cases show a distance of
greater than one-fourth mile between the VLS site and the BR site
(67.3%) and between the IC site and BR site (65.1%). Table 8
summarizes the key distances between murder incident compo-
nent pairings.
It was predicted that when the distance between locations
or a pair of locations is less than 199 ft, the relatively close
proximity of the locations will enhance the case solvability
of a murder of an abducted child; the percentage of cases
solved will be significantly greater than when the locations are
separated by more than 199 ft—only the distance interval less
than 199 ft between the MS/BR site pairing indicates a signi-
ficant effect on solvability in the direction predicted. In general,
the results from Kendall’s Tau-b Test do not support the
prediction set out by this hypothesis. The results are shown in
Table 9.
Time and Distance Interaction Effect on Case Solvability
It was predicted that when the time between a pair of locations
is greater than 24 h and the distance between the same pair is
greater than 199 ft, such a relatively distant proximity in time and
distance will not contribute to the case solvability of a murder of
an abducted child; the solvability will diminish sharply in cases
when both the time span and distance interval are shorter for that
pair of locations. Because of insufficient frequencies in the time
interval categories in the other murder incident component pair-
ings, only the following murder incident location pairings were
examined for this hypothesis:
1. VLS site/BR site,
2. IC site/BR site,
3. MS/BR site.
A Discriminant Analysis was performed on relationship of time
less than or greater than 24 h and distance less than or greater than
199 ft to case solvability for each pairing.
Because one of the independent variables was eliminated from
the model by the discriminant analysis for each murder incident
location pairing examined, the hypothesis that time spans greater
than 24 h and distances greater than 199 ft enhance case solva-
bility is not supported. In addition, it was hypothesized that the
solvability would decrease significantly if time and distance
proximity were shorter for each murder incident location pairing.
This hypothesis is not supported for any murder incident location
pairing in child abduction murder investigations.
Discussion
Results show that when any information on the dates and
locations of the murder incident components is known, the
probability of child abduction murder case solution increases.
There is a strong positive correlation between knowing the time of
a murder incident component and the ability to identify a
perpetrator. The strongest indicator of solvability, when investi-
gators know the time of a murder incident component, is the MS.
The IC site also contributes strongly to solvability when informa-
tion about the time is known. Unfortunately, little attention is
given in homicide investigation courses on how to locate and
process this site even though knowing the actual time of the initial
contact between the victim and offender or murder increases the
ability of investigators to verify or refute an offender’s alibi.
This research also shows that in child abduction murder cases,
shorter time proximity between murder incident locations has no
significant impact on case solvability. However, there is support
that case solvability will increase when the MS/BR site murder
incident component pairing is separated by more than 24 h. This
refutes the general belief that when an offender separates the time
between the murder and when he disposes of the victim’s body
that case solvability will decrease. However, confounding factors
associated with child abduction murders could account for this
finding such as: the extensive resources provided to child abduc-
tion murder investigations which would not be expended in
general homicide investigations or the use cold case teams.
Previous solvability research shows that the more investigators
know about the distances between the pairs of the murder incident
components, the more case solvability will increase; this study
shows similar findings. When the distance between the MS and
the BR site locations is less than 199 ft, solvability increases. A
reason for this finding could be that the murder occurred where the
body was recovered, resulting in a greater amount of physical
evidence discovered in one location linking the offender to the
crime. Distance less than 199 ft between the remaining five
TABLE 7—Distance between murder incident components.
Distance interval
Percent of cases
VLS
to IC
VLS
to MS
VLS
to BR
IC
to MS
IC
to BR
MS
to BR
0–199 ft 64.2 26.6 15.8 34.0 19.7 69.1
200 ft–1/4 mile 15.9 19.2 16.8 14.3 15.2 6.1
41/4–1 12 miles 8.3 13.7 13.4 11.5 11.6 4.5
41 1
2–12 miles 7.2 24.5 30.5 24.4 28.7 10.8
412 miles 4.4 16.0 23.4 15.9 24.8 9.6
VLS, victim last seen; IC, initial contact; MS, murder site; BR, body
recovery.
TABLE 8—Key distances between murder incident components.
Initial Contact
Site
Murder
Site
Body Recovery
Site
Victim last seen site
199 41/4 mile 41/4 mile
(64.2%) (54.2%) (67.3%)
Initial contact site 41/4 mile 41/4 mile
(51.8%) (65.1%)
Murder site
199 ft
(69.1%)
TABLE 9—Relationship between distance and case solvability.
Murder Incident Component
Pairing Distance
Percent of
Cases Solved n Tau b p
Victim last seen site to initial
contact site
199 77.6 177 0.039 0.329
4199 80.9 331
Victim last seen site to
murder Site
199 78.9 352 0.099 0.007
4199 87.7 142
Victim last seen site to body
recovery site
199 71.4 421 0.087 0.011
4199 82.0 91
Initial contact site to murder
site
199 85.2 323 0.021 0.612
4199 83.6 163
Initial contact site to body
recovery site
199 79.7 408 0.011 0.784
4199 78.6 99
Murder site to body recovery
site
199 88.0 162 0.105 0.005
4199 79.3 326
BROWN AND KEPPEL . CHILD ABDUCTION MURDERS: TIME AND DISTANCE SEPARATION EFFECT ON CASE SOLVABILITY 143
murder incident component locations has no significant impact on
case solvability.
Relatively close time and distance proximity between murder
incident component pairs does not contribute significantly to case
solvability. In addition, when the time and distances proximity
decreases among pairs of murder incident components, the rela-
tively distant proximity in time and distance does not contribute to
case solvability.
Investigative Implications
These results have important investigative implications. This
study is a valuable tool for use in murder investigations of
abducted children because the findings add to the understanding
of these investigations and indicate which locations are critical to
case solvability. Because time and distance do not play the same
role in case solvability in CAM investigations as in general
murder investigations, there may be other factors, such as type
of forensic evidence, investigative resources, or actions by first
responders which impact case solvability.
Because child abduction murder investigations are draining on
department resources, further research on prioritization of inves-
tigation protocols and personnel resources to maximize case
solvability is necessary. Fiscal considerations greatly influence
investigations, therefore additional research should also be con-
ducted on the resources an agency expends on these types of
investigations and their commitment to that resource allocation for
the duration of the investigation. In addition, attention should be
given to the budgetary process of resource allocation by the
appropriate governmental bodies responsible for law enforcement
agency budgets.
Research should also be conducted on the types of physical
evidence left by the victim or offender during an abduction
murder and the impact of that evidence on case solvability.
Empirically supported research in this area will enable agencies
to focus limited police resources on those items of evidence most
likely to influence case solvability. Because additional or different
factors may affect whether or not an identified offender is
convicted, a closer examination of factors relating to offender
conviction should also be undertaken. Finally, a comparison
between abduction cases in which a child was recovered alive
and those in which a child was murdered would provide valu-
able—perhaps even lifesaving—information.
Because the occurrence of child abduction murder is rare, it is
likely that most jurisdictions will not have experience in dealing
with this type of investigation. This research study provides
valuable, empirically supported, information to homicide detec-
tives investigating the murder of an abducted child. The informa-
tion contained within this study will enable police to efficiently
allocate resources to the investigation. Investigators can use this
information to prioritize how they pursue leads which relate to
each murder incident component according to their influence on
case solvability. In addition, the results of this study can be used to
develop training materials for law enforcement personnel.
Conclusion
Given the effect of extreme media coverage of child abduction
murders, and the intense pressure from victim’s advocacy groups,
it is surprising that little empirical research has been undertaken
before now to determine the effect of time and distance on case
solvability. This research study increases the body of knowledge
relating to time and distance as solvability factors in child
abduction murder investigations. The findings indicate that in-
formation about the time and distance are critical solvability
factors in murder investigations of abducted children. It is
imperative that CAM be further explored in order to give police
a larger arsenal of investigative tools and parameters for these
types of investigations. The information obtained from this study
is also a valuable community resource which will enable the
public to better understand the risk of victimization from a child
abduction murder and the complex nature of these types of
investigations.
The discovery that information about time and distance in-
creases case solvability is hardly unexpected. This research
quantified what experienced homicide detectives already know.
However, the findings about the nature of the time and distance
intervals between murder incident component pairings and their
effect on solvability indicate that time and distance relationships
operate in a unique manner in murders of abducted children. As
tragic as the death of any abducted child is, the greater tragedy
would be a case that remained uncleared because local law
enforcement investigators were unaware of major factors which
would have increased case solvability.
Acknowledgments
The authors wish to thank Scott Vollum, Ph.D., for his help
with data cleaning and coding and Kc L. Wendler Deaver, M.Sc.,
for her help with data verification. In addition, the authors wish to
thank Charles M. Friel, Ph.D., and Jerry L. Dowling, J.D., for their
contributions to the original study as part of Ms. Brown’s thesis
committee. Finally, the authors wish to thank the anonymous
reviewers for their careful review and valuable suggestions in
clarifying this article.
References
1. National Center for Missing & Exploited Children. Missing and abducted
children: a law-enforcement guide to case investigation and program
management. Alexandria, VA: The Center, 2000.
2. Hanfland KA, Keppel RD, Weis JG. Investigative case management for
missing children homicides. (Cooperative Agreement 93-MC-CX-K006),
Attorney General of Washington, Olympia, WA, 1997
3. National Incidence Studies of Missing, Abducted, Runaway, and Thrown-
away Children in America (NISMART). Highlights from NISMART
bulletins. Washington, DC: U.S. Department of Justice, Office of Juvenile
Justice Programs, Office of Juvenile Justice and Delinquency Prevention,
2002.
4. Brown KM. An analysis of the effect of time and distance relationships on
case solvability in murder investigations of abducted children. Master’s
thesis, Sam Houston State University Huntsville, TX, 2005.
5. Beyer KR, Beasley JO. Nonfamily child abductors who murder their
victims. J Interpers Violence 2003;18(10):1167–88.
6. Boudreaux MC, Lord WD, Dutra RL. Child abduction: aged-based
analysis of offender, victim and offense characteristics in 550 cases of
alleged child disappearance. J Forensic Sci 1999;44(3):539–53.
7. Egger SA. The killers among us: an examination of serial murder and its
investigation. 2nd ed. Upper Saddle River, NJ: Prentice Hall, 2002.
8. Keppel RD, Birnes WJ. The psychology of serial killer investigations: the
grisly business unit. San Diego, CA: Academic Press, 2003.
9. Greenwood P, Petersilia J, Chaiken J. The criminal investigation process.
Lexington, MA: D.C. Heath, 1977.
10. Danto BL, Bruhns J, Kutscher AH. The human side of homicide. New
York: Columbia University Press, 1982.
11. Lunde DT. Murder and madness. Stanford, CA: Stanford Alumni Associa-
tion, 1975.
12. Geberth VJ. Sex-related homicide and death investigation: practical and
clinical perspectives. New York: CRC Publishing, 2003.
13. Geberth VJ. Practical homicide investigation: tactics procedures and
forensic techniques. 3rd ed. New York: CRC Publishing, 1983.
14. Adelson L. The pathology of homicide. Springfield, IL: Charles C.
Thomas, 1974.
144 JOURNAL OF FORENSIC SCIENCES
15. Spitz W, Fisher R. Medicolegal investigation of death: guidelines for the
application of pathology to crime investigation. 3rd ed. Springfield, IL:
Charles C. Thomas, 1993.
16. Fisher B, Svenson A, Wendel O. Techniques of crime scene investigation.
New York: Elsevier Publishing Co, 1986.
17. Ressler RK, Burgess AW, Douglas JE. Sexual homicide: patterns and
motives. New York: Lexington Books, 1988.
18. Mott NL. Serial murder: Patterns in unsolved cases. Homicide Stud
1999;3(3):241–55.
19. Sweet RW. Missing children: found facts. NIJ Rep 1990;222:15–8.
20. Keppel RD, Weis J. Time and distance as solvability factors in murder
cases. J Forensic Sci 1993;39(2):286–401.
21. Keppel RD. An analysis of the effect of time and distance relationships in
murder investigations [dissertation]. Seattle, WA: University of Washing-
ton, 1992.
Additional information and reprint requests:
Katherine M. Brown, M.A.
College of Criminal Justice
Sam Houston State University
Huntsville, TX 77341-2296
E-mail: kbrown@shsu.edu
BROWN AND KEPPEL . CHILD ABDUCTION MURDERS: TIME AND DISTANCE SEPARATION EFFECT ON CASE SOLVABILITY 145
=