Science and Justice 55 (2015) 509–513

Contents lists available at ScienceDirect

Science and Justice

journal homepage: www.elsevier.com/locate/scijus

Utility of the clue — From assessing the investigative contribution of

forensic science to supporting the decision to use traces
Sonja Bitzer a,⁎, Nicola Albertini b, Eric Lock c, Olivier Ribaux a, Olivier Delémont a
a
Ecole des Sciences Criminelles, Université de Lausanne, Batochime, 1015 Lausanne, Switzerland
b
Police Cantonale Vaudoise, Service d'Identité Judiciaire, Centre Blécherette, 1014 Lausanne, Switzerland
c
Police Cantonale Genevoise, Brigade de Police Technique et Scientifique, Boulevard Carl-Vogt 17-19, 1205 Geneva, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: In an attempt to grasp the effectiveness of forensic science in the criminal justice process, a number of studies in-
Received 17 October 2014 troduced some form of performance indicator. However, most of these indicators suffer from different weak-
Received in revised form 17 April 2015 nesses, from the definition of forensic science itself to problems of reliability and validity. We suggest the
Accepted 22 May 2015
introduction of the concept of utility of the clue as an internal evaluation indicator of forensic science in the in-
vestigation. Utility of the clue is defined as added value of information, gained by the use of traces. This concept
Keywords:
Forensic science
could be used to assess the contribution of the trace in the context of the case. By extension, a second application
Utility of this concept is suggested. By formalising and considering, a priori, the perceived utility of using traces, we in-
Effectiveness troduce the notion of expected utility that could be used as decision factor when choosing which traces to use,
Impact once they have been collected at the crime scene or from an object in the laboratory. In a case-based approach,
Investigation utility can be assessed in the light of the available information to evaluate the investigative contribution of foren-
sic science. In the decision-making process, the projection or estimation of the utility of the clue is proposed to be
a factor to take into account when triaging the set of traces.
© 2015 The Chartered Society of Forensic Sciences. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction
Forensic science, being a relatively young discipline, has developed
substantially in the past decades. Due to the increase in DNA and drug
testing, forensic science laboratories have multiplied, university
programmes continue to proliferate and technical developments con-
tinue to attract most of the attention. At the same time, the broad
media attention shaped the view of forensic science as the omnipotent
tool to help solve crime affecting not only the understanding of the fo-
rensic science role in the overall population, but also the everyday prac-
tice of forensic scientists and the expectations in courtrooms. This
combines to shape a perception of forensic science as being an infallible
science [1].
Fundamentally opposed to this view is the depiction in recent
studies, challenging forensic science in its nature (e.g. [2]) and the effec-
tiveness of its contribution to the criminal justice process [3,4]. In re-
sponse to this commotion, several studies attempted to measure the
efficiency or effectiveness of forensic science using a range of different
indicators, essentially considering an economical, sociological or polic-
ing perspective (e.g. [5–7]). However, these indicators adopt an external
⁎ Corresponding author. Tel.: +41 21 692 46 21; fax: +41 21 692 46 05.
E-mail address: sonja.bitzer@unil.ch (S. Bitzer).
perspective of the criminal justice process. Albeit their focus is on one
precise aspect of the contribution of forensic science, the responses of
theses indicators are extrapolated to the global impact of forensic sci-
ence in the criminal justice process, providing an unbalanced view.
We suggest the introduction of the concept of utility of the clue to
evaluate the contribution of traces to the investigation. The latter, de-
fined as added value of information attainable by the use of the trace,
depends on the informative potential of the clue itself, as well as the in-
formation available before the selection of traces. The aim is to depict
the contribution of forensic science in different chapters of the criminal
justice process and to suggest an indicator allowing to assess the added
value of information provided by used traces.
By extension, we propose to use utility of the clue in the decision to
use traces as an aid to make more appropriate, conscious and targeted
triaging decisions. Once traces have been collected at the crime scene,
a decision has to be made about which traces are used. The factors in-
volved in this decision are not clearly identified or explicitly stated in
the corresponding literature.
In the following section the current state of the literature regarding
performance indicators, their connection with the definition of forensic
science and the factors involved in the decision to use traces will be
discussed. In Section 3, the concept of utility in general and utility of
the clue will be presented and discussed in the context of performance

http://dx.doi.org/10.1016/j.scijus.2015.05.005
1355-0306/© 2015 The Chartered Society of Forensic Sciences. Published by Elsevier Ireland Ltd. All rights reserved.
510 S. Bitzer et al. / Science and Justice 55 (2015) 509–513
measurement and decision making processes. Finally, a case example is
presented to illustrate the application of the suggested concept.
2. Literature review
In early studies, police use of forensic science, in terms of number of
scenes visited and number of traces collected and submitted by scene of
crime officers, was scrutinised. Albeit variable between different police
forces [8], the use of forensic science was found to be consistently low
[9–12]. In 2009, Schroeder and White [13] reinforced this view in their
study of the utilisation of biological samples in homicide cases from
1996 to 2003. The Pathfinder Project [14] and the DNA Expansion Pro-
gramme [15] tried to address precisely this problem by promoting the
use of traces as a mean to improve policing. However, their conclusions
were contradictory, with one stating that an increased use was influenc-
ing the number of identifications and the other one stating that the in-
crease did not measure an effect on case outcomes. Adderley et al.
[16] found that an increase in the number of collected biological traces
would not “dilute the matching rates for DNA samples” [16, p.176]
and, similarly, Roman et al. [5] determined that an increased use of
rapid DNA for burglary cases allowed to increase the number of suspect
identifications.
Another way of attempting to measure the effectiveness of forensic
science and its databases is by looking at the possible increase in the
number of identifications [14]. While this factor appears to be pertinent
at first, especially when comparing to the invested resources (e.g. num-
ber of scenes visited), it constitutes an oversimplified approach to mea-
sure the value of forensic science. From the decision to send a crime
scene investigator to scene to the identification of a suspect by traces,
multiple decisions need to be made: which scenes are investigated by
the crime scene investigator, which traces are collected, which traces
are analysed, which techniques are used in the laboratory to analyse
the traces, are the results of the analysis sent to the national database,
etc. When considering the aforementioned ratio, all these decision
steps are reduced to one measure and confound different indicators of
effectiveness on different stages of the decision making process.
As the infrequent use of forensic science was perceived as being
problematic, its reasons were examined. Generally, several explanations
were proposed, including the lack of knowledge of police investigators
when it comes to the potential value or utility of traces for their
case [12,17–19], which could possibly be comprehensible in the
late 1980s or the early 1990s (due to the recent development of DNA
analysis). However, by 2009, this conclusion still found support [20],
which is more surprising considering the large publicity of forensic sci-
ence and the strong result focused culture of police officers (i.e. investi-
gators rapidly integrate new techniques that deliver decisive results).
Nevertheless, this lack of knowledge might not have had the same
reasons at this point in time. Inadequate communication, already men-
tioned in 1987, might be at the core of the problem [17]. The infrequent
use might be caused by a lack of awareness of the available techniques
on the side of the police, and a lack of awareness of the available traces
on the part of the laboratory scientists. Such inadequacy in communica-
tion may be favoured by the model of integration of forensic science
practice in the enquiry. As described by Williams [21], a “structural” dis-
sociation of forensic science from police may enhance this knowledge
gap. An extreme picture that could be drawn is that scientists, separated
from police investigation, have to analyse what police investigators ask
them to. Hence, the decision to use traces with all the dimensions in-
volved in this decision (which traces, when, how, by which means
and why) lies essentially with police investigators. This lack of commu-
nication and knowledge might therefore cause the tendency for the use
of traces as means of last resort [13,22].
All the latter concerns may be explained by a misconception of the
roles and responsibilities of the different actors in the criminal justice
system. Laboratory scientists perceive crime scene investigators as
mere evidence collectors (as opposed to forensic investigators or
specialist advisor), without following a reasoning process or selection
criteria, and hence, do not acknowledge their substantial contribution
to the complete process [23]. Forensic scientists can assume several dif-
ferent roles corresponding to differentiated needs imposed by the struc-
tural or procedural constraints. Basically, the role is threefold: (1) the
investigation needs to be carried out and the relevant information
found, (2) the information needs to be structured and finally (3) the
traces need to be considered in the light of competing hypotheses of in-
terest [24]. Furthermore, the criminal phenomenon at hand needs to be
considered and understood and the gained information integrated into
the investigation process. This may be the definition of the roles, consid-
ering the “procedural” integration model of forensic science and police
investigation described by Williams [21]. This “expert collaboration”
approach recognises and acknowledges the utility of expert knowledge
and the potential insufficiency of knowledge on the side of the users of
the service to offer informed assessments of its quality and its investiga-
tive potential [21, p.772]. The roles and competences diverge in regard
to the integration system chosen by institutions and organisations [21,
25]. In addition, this dissociation of the roles is accompanied by a hier-
archy established between the investigator and the scientist.
The commonly understood role of forensic science as the assistance
of court through the application of scientific techniques in an adversar-
ial system is prone to dramatically understate the potential of traces as it
focuses on the judicial phase only. It epitomises forensic science as the
analysis of the trace and the interpretation of the results, without con-
sidering the reasoning processes and all the decisions involved. Most
of these studies deal with the anglo–saxon system, or more precisely
with the British system, which has been influenced by the Home Office
and its policing strategies. In particular, the roles and responsibilities of
the actors in the system change depending on the system at hand. These
concerns, the controversies in the definition of forensic science, the in-
tegration model as well as the use of purely quantitative, narrowed per-
formance indicators (“forensic yield”, “forensic matches”, lack of validity
and reliability), have already been raised by Williams and Weetman
[26] when they described the weaknesses of the current measurement
of support of forensic science to homicide investigation in the UK. The
definition of forensic science and the integration model chosen are
intertwined subjects and as such, are difficult to be changed. However,
the third weakness mentioned is the choice of the performance
indicator.
In order to measure the effectiveness of forensic science, some au-
thors concentrated on the predictive power of trace processing on
case outcomes. Although the proportion of submitted traces was close
to 100% for homicide cases, their influence on any of the stages of the
process (arrest, referral, charging, conviction) remained ambiguous/
undefined; two groups of authors working with the same dataset ob-
tained differing results and reached different conclusions [3,7]. Hence,
the results and conclusions seem to be more influenced by the chosen
methodology than by the data itself. Other factors, such as case charac-
teristics, seem to be involved in case clearance. An interesting result was
that most robberies went unsolved due to a lack of witnesses [3]. Hence,
case circumstances and police strategies (in finding and questioning
witnesses) seem to be the most influential factor regarding the detec-
tion of robbery cases. Similar results were found for homicide
cases [4,27–29]. All these studies determined that police actions or the
offender's attitude were indicators for homicide clearance. In the
study performed by Brodeur [4], scientific evidence helped to arrest a
suspect in a mere 2% of cases. Whereas, in the study performed by
Mucchielli [28], in only 7% of cases, physical traces collected from
crime scenes were the crucial element leading to the offender(s).
These findings conducted Mucchielli to emphasise the determining
role of these traces, noting that they had played only a secondary role
in other cases. He stated further that physical traces were more often
than not getting a value once the scenario had been established, so
their role would be again indirect. This would confirm the essential
yet not determinative role of the work on the crime scene.
 S. Bitzer et al. / Science and Justice 55 (2015) 509–513
The significance of a potential influence of forensic science on the
criminal justice process is still vague, as it appeared to be the mere pres-
ence of traces (with examination taking place after arrest) being predic-
tive for arrest. The question arises if traces are only collected and
analysed once a suspect is apprehended (direct arrest, e.g. Baskin and
Sommers [30]). Thus, arriving at the conclusion that traces are collected
and analysed in order to build the case against the suspect [17,18]. This
would also mean that if no suspect is immediately found, investigators
do not see the need to collect or proceed to the analysis of traces for
the progression of the case [3]. Similarly, in studies examining the
unanalysed traces in rape and property crime cases [31,32], the main
reason for the non-submission of traces was stated as being the lack of
a suspect in the case. However, a large proportion of agencies responded
also that they were not sure about the usefulness of forensic evidence
(in their case). This result can be interpreted in two ways: either this
constitutes a lack of knowledge of the investigators about the usefulness
of traces in general, or they lucidly (and maybe correctly) assessed the
expected utility of the clues in their case, which led to the decision not
to submit traces for examination. In general, the question arises on
what grounds the analysed traces are chosen amongst the collected
ones. Furthermore, it remains undetermined if the not submitted traces
are a consequence of an active choice for other traces (e.g. due to a good
understanding of time constraints or preference for other traces) or are
consciously rejected, for lack of usefulness to their case for instance.
Traces seem to be preferentially chosen by (internal) managerial or
actuarial rules or political pressure (strategic dimension as proposed by
Ribaux et al. [33]), as opposed to the consideration of their utility. The
main reasons to analyse a trace are the urgency of the case, the serious-
ness of the offence, the presence of a suspect and the prosecutor or judi-
cial request [34,35]. However, the simple facilitation of the use does not
necessarily lead to a more appropriate use of traces. Hence, the consid-
eration of utility of the clue could lead to a more conscious and concert-
ed use of traces, and thus, to a rationalisation in resource consumption.
Indeed, a pilot project carried out in Belgium showed that in cases
where a forensic consultant was present during the reunions between
the prosecutor, investigators and analysts, the choice of traces was
more effective and efficient. The forensic consultant's function was to
advice on the choice of traces to be analysed regarding the pertinence
or, better, their utility in the case [36].
3. Utility of the clue
Two concepts related to the measurement of the contribution or im-
pact of a discipline in a (decision-making) process are efficiency and ef-
fectiveness. If efficiency corresponds to doing things right and
effectiveness to doing the right things, what we would really be looking
for is the combination of both: doing the right thing right. This brings
us to utility and expected utility, which have long been used in decision
theory [37,38]. The definition used in the context of this study is based
on Soergel's definition in information theory [39]. An entity is consid-
ered topically relevant for a question, when it sheds light on the latter.
It is considered pertinent, if it is “relevant and if it is appropriate for
the person, that is, if the person can understand the document and
apply the information gained. […] An entity has utility if it is pertinent
and makes a useful contribution beyond what the user knew already”
(see [39], p.590).
Before adapting this concept to the context of forensic science, the
difference between the notions of trace, clue and evidence used in this
study needs to be clarified. Traces are silent witnesses, which are
vectors of information. This information is not directly intelligible or
seizable, it needs to be perceived and considered in the context of the
different elements such as the case, the background and the surface
on which the trace is left. Traces are ephemeral remnants of past events,
like fading symptoms of an illness, whose causes remain to be clarified.
The information gained through these traces – the clues – are at the
basis of the reconstruction process of past events, shape it, with more
511
or less reliability and certainty. Evidence is another state of this same
object, but its consideration by the judge for court purposes makes it
measurable and a link to the case is required [40].
For a clue to be considered useful, the information retrieved from the
trace – the clue – must fulfil three conditions, analogously to the con-
cept of utility defined by Soergel [39]. Not only does it (1) have to be re-
lated to the case in question, and (2) be understood by its user (e.g.
investigator), but also, (3) it has to shed light on a question in the
case, without being redundant. The clue needs to add new and valuable
information to the case, which contribute to the processing of the case.
It is this second part which is important to consider. The fact that the in-
formation is new will not necessarily bring more information to the case
(e.g. processing a fingermark left by a single offender when its identity
has already been confirmed by the analysis and comparison of a biolog-
ical trace is redundant when considering the sole identification poten-
tial of both traces). In addition, this information needs to be gained in
an efficient way, without redundancies (at best). Thus, utility of the
clue in the context of each specific case is dependent on the information
potential of the trace itself, but simultaneously it is contingent on the in-
formation already available, through other analyses or the investigation
itself for instance.
3.1. Performance indicator for forensic science
Considering the performance indicators that are currently used to
measure the performance, the infrequent use of forensic science is con-
sidered as the manifestation of a low performance of the investigators
regarding the use of traces [41].
However, a correct appreciation of the usefulness of clues to the case
might appropriately limit or redistribute their use in the investigation.
As this factor has rarely been taken into account when studying the
use of forensic science and their reasons, it remains, thus far, only a con-
jecture. Notwithstanding, the limited definition of forensic science that
is often promoted (leading to its use for court purposes only) demon-
strates the incomplete understanding of the diversity of usefulness of
traces in the criminal inquiry, even in serious crimes [42].
By 1996, the Association of Chief Police Officers [43] already sug-
gested to assess the contribution of forensic science from the inside by
considering the usefulness of the analysis as performance indicator for
forensic science in the inquiry. Although this concept is closely linked
to the utility of the clue, it is limited by the fact that it assesses the use-
fulness of the result, as opposed to the utility of the information re-
trieved from that result.
In their 5-year project, Julian et al. [44] attempted to “identify when,
where and how forensic science can add value to criminal investiga-
tions, court trials and justice outcomes” [44, p.217]. Indeed, the useful-
ness of forensic science is much more diverse than solely for
identification purposes or production of evidence. None of the studies
accounted for the number of cases in which clues, provided by the use
of traces, guided the police interrogation of the suspect, for instance. Al-
though this utility is indirect, it is of great importance to take into ac-
count when assessing the beneficial contribution of forensic science to
the overall criminal justice process. Williams and Weetman [26] refer
to this concept as “contextual utilities”, and which measurement re-
quires the “consideration of the necessary, typical and exceptional
achievements” at each of the stages of the judicial process “as well as
the practical contingencies to which they are subject” [26, pp.383–384].
The implementation of utility of the clue as performance indicator
can also be seen as the utility a posteriori of the used traces, once the
analyses have been performed, their results are known, context infor-
mation has been clearly determined (e.g. number of suspects), etc. An-
other aspect of performance measurement is the consideration of the
expected utility (projection) of the non-used traces, without knowledge
of their analysis outcome, in comparison with the expected utility of the
used traces. The possibility, by extension, to evaluate the expected util-
ity before analysis that could be useful in the decision-making process
512 S. Bitzer et al. / Science and Justice 55 (2015) 509–513
regarding the use of traces will be further discussed in the following
section.
3.2. Decision factor in the decision to use traces
In the processing of traces, from crime scene to court, several deci-
sion steps are encountered leading to substantial triaging and attrition
of traces along the process. At first, the trace, once it has been detected,
needs to be collected. This first step depends on multiple factors such as
the quality of the traces on the crime scene or objects judged by the
crime scene investigator and is probably influenced by the anticipated
utility of this clue to the case. Then, the decision to use traces in the in-
quiry comes into play, with a wide understanding of the term use: from
the use of the information retrievable from the trace as such (without
analysis), to the analysis and the use of the results after analysis. Finally,
the decision to use traces for the production of evidence to court arises,
but is of lesser interest in the scope of this study.
While collecting large amounts of traces at crime scenes would be
understandable, investigators then have to make a choice of traces to
be further exploited. It is difficult to defend the decision to limit the
number of items or traces to be collected at the crime scene, as conse-
quently, these traces would not be available for use and analysis. How-
ever, it is likely that the decision to collect certain traces is already
guided by the anticipation of the traces to be used later in the enquiry.
The other – obvious – relation between the decision to collect traces
and the decision to use them is the chronological one: only what is
available can be used. An available and limited set of traces ensues as
the result of the decision to collect certain traces but not others and in-
fluences the decision to use them.
The triaging decision is especially important when handling large
amounts of traces with limited resources. As the question of efficiency
and effectiveness of forensic science is increasingly of interest due to fi-
nancial pressure, this decision phase becomes the focal point of the ma-
terial trace processing. Although the decision to use a trace in order to
extract information from it has never been the subject of a study thus
far, it is, together with the factors leading to it, of utmost importance re-
garding the usefulness of forensic science in the investigation. This deci-
sion includes several factors such as which traces are chosen to be used,
who makes this decision, at what stage in the inquiry process and, why.
As such, the utility of the clue is dependent on the roles and responsibil-
ities of the actors in the system. Due to the differing access to informa-
tion, the perceived utility of the clue can change. Furthermore,
external factors, similar to the strategic, physical, immediate and crimi-
nal environments of the case as proposed by Ribaux et al. [33] for the de-
cision to attend and how to search a crime scene, are also influencing
this decision.
The utility of a clue can thus be manifold and depends on several
factors. The principal and most obvious utility is, of course,
individualisation or corroboration of a suspect. However, in many
cases, it can be important to consider also the intelligence component
of the information or the reconstruction of the micro sequence of the
events [20,45]. In this same direction, Tilley and Townsley [46] sug-
gested an “organic” integration model. They emphasised that the “re-
duction of volume crime, the quick elimination of the innocent and
the conviction of the guilty” are optimal contributions of forensic sci-
ence and need to be reached by “understanding and managing the
human, social side of the forensic process” [46, p.376].
The utility of the clue can be put on a same level with the aims of its
uses, as it is the use of the trace that may lead to a gain new information.
One example of this utility is for the provision of intelligence: “In addi-
tion [to be decisive or at least relevant], even where such artefacts are
assumed not be usable for the construction of a case against an actual
or potential suspect for a particular crime, scene examiners still have
to consider whether or not their construction and analysis may contrib-
ute to the collation of police intelligence relevant to current or future in-
vestigations of other crimes” [21, pp.763–764]. The reasons why a trace
is used refers to the question of what information one wants or needs to
gain through the use of this trace. The expected utility of the clue needs
to be considered in order to understand why some traces are chosen to
be used over others. At the same time, the decision to use a trace is in-
fluenced by factors facilitating or impeding the use of a trace. By exten-
sion, the decision not to use a trace is also of great interest as it remains
unclear if the traces, that are not used, are the mere consequence of the
decision for other traces or if it is deliberately chosen not to use these
traces.
How utility of the clue is implemented in a decision-making process
in the investigation remains to be clarified. Many factors influence the
perception of the utility of the various clues, like for instance, the inves-
tigators themselves, the subjective judgement of the quality of the
traces, etc.
3.3. Case example
An example is given to showcase the utility of the clues in the case,
and their changing nature depending on the available information. It
mimics, in a simplified manner, a combination of situations of real cases.
Let's consider a robbery case (one offender seen on surveillance
camera images) with one shoe mark and two biological traces collected
at the crime scene. The general expected utility of the shoe mark is to
link cases with each other. One biological trace is analysed to identify
the offender The decision for one of the two biological traces was
made on the basis of the quality of the trace, the matrix, the pertinence
of the trace, etc. The analysis of the general pattern of the shoe mark
permits, already at the crime scene, to give information to the enquiry
in terms of shoe brand. The course of the offender does not need to be
reconstructed by the shoe marks, as the surveillance camera images
are available and already provide a detailed sequence of the offender's
actions. This sequence of actions can help choose pertinent traces,
which can generally be considered a form of utility of the shoe marks.
The result of the analysis of the biological trace is positive and a pro-
file is yielded (different from that of the victim). However, no DNA
match could be provided when comparing with the National DNA
database. Hence, the utility a posteriori is null, except for its exclusion
potential. The expected utility of the not analysed biological trace is
low, as the profile already available can reasonably be attributed to
the offender due to the high pertinence of the trace. If the pertinence
would not be given, it could still be reasonable to wait until the result
of the first analysis is obtained before analysing a second trace (to pre-
vent redundancy).
In a second robbery case, the offender is arrested immediately after
the offence, due to police investigation. In the audition, the offender de-
nies any implication in other cases. At the second crime scene, one shoe
mark and one biological trace are collected. The analysis and compari-
son of the shoe mark reveal a possible link with the first robbery case.
The biological trace is not analysed. However, a buccal swab of the
offender is performed and analysed. A DNA match with the biological
trace from the first robbery case is yielded. Confronted with this result
during audition, the offender confesses to having committed the first
robbery, too. The utility a posteriori of this biological trace changes
then to aid in the inquiry/audition and confirmation of the implication
of the offender in the case. The not analysed biological trace from the
second robbery has an expected utility of confirmation of the implica-
tion of the offender (validation of the confession and corroboration of
the possible link unveiled by the shoe marks).
4. Conclusion
Effectiveness and efficiency measures used thus far have shown to
present several weaknesses which reveal a limited understanding of
the manifold nature and role of forensic science and an outside assess-
ment of the contribution of forensic science to the inquiry (i.e. econom-
ical, sociological, or criminological view in an actuarial perspective (flow
 S. Bitzer et al. / Science and Justice 55 (2015) 509–513
analysis)). The introduction of the concept of utility of the clue, defined
as added value of information obtained by the use of traces, is suggested
as performance indicator of forensic science in the investigation in order
to evaluate its contribution from an internal, clinical perspective, grasp-
ing the whole variety of its contribution. This concept is intrinsic to the
traces themselves, it depends on their information potential in the spe-
cific situation. In addition, it depends on the other information available
at the time of decision-making. In this case-based approach, the utility
can be assessed in the light of the available information, once the
analyses have been carried out. The actuarial approach proposed in
earlier studies allows for a global view of the use of forensic science
through means of statistics, where a single (or group of predefined)
indicator(s) is measured for a whole set of cases. However, a more clin-
ical, case-based approach would allow to capture the whole variety of
the different uses of the traces in order to measure the real contribution
of forensic science to the judicial system. Through a detailed study of the
decision-making process in specific cases, a picture of the contribution
of clues to the inquiry can be drawn and might be generalised by recur-
ring utilisation schemes. The suggested methodology is a case-based
approach by observing ongoing cases in the investigation phase. Fur-
thermore, the information obtained can be completed by interviews
with key actors in the criminal justice system.
By extension, an application of this concept can also be conceived in
the decision to analyse traces, once they have been collected at the
crime scene. The anticipation or estimation of their utility is proposed
to be a factor to take into account when triaging the set of traces. How-
ever, the expected utility is assessed with more difficulty, as many fac-
tors still remain unclear (such as the number of suspects for instance).
Furthermore, depending on the other information available at the
time of deciding which traces to use, the added value of information
gained through the use (analysis and further exploitation) of traces
can be balanced against each other and thus, a more informed, con-
scious and focused decision could be made.
Novelty statement
The contribution of forensic science in the judicial practice is cur-
rently being scrutinised. Many indicators have been considered in this
perspective trying to assess the contribution of forensic science. These
indicators are mainly oriented to measure the performance or efficiency
of forensic science, primarily in the perspective of the production of ev-
idence for court decisions. We argue that such indicators do not suitably
capture all levels of the diverse contributions of forensic science in the
criminal justice system in general and the investigation in particular.
Hence, we suggest the introduction of an original concept, the utility
of the clue, as an internal assessment factor. Furthermore, a second ap-
plication is provided. Utility of the clue could be used as guiding factor
in the decision making process when triaging traces in the investigative
phase.
Acknowledgments
The authors would like to thank the anonymous reviewers for their
valuable comments that helped improve the quality of the article.
References
[1] C. Roux, R. Julian, S. Kelty, O. Ribaux, Forensic science effectiveness, in: G. Bruinsma,
D. Weisburd (Eds.), Encyclopedia of Criminology and Criminal Justice, Springer
2013, pp. 1795–1805.
[2] National Academy of Sciences, Strengthening Forensic Science in the United States:
A Path Forward, National Research Council, 2009.
[3] D. Baskin, I. Sommers, The influence of forensic evidence on the case outcomes of
homicide incidents, J. Crim. Just. 38 (2010) 1141–1149.
[4] J.-P. Brodeur, L'enquête criminelle, Criminologie 38 (2005) 39–64.
[5] J. Roman, S. Reid, J. Reid, A. Chaflin, W. Adams, C. Knight, The DNA Field Experiment:
Cost-effectiveness Analysis of the Use of DNA in the Investigation of High-volume
Crimes, Urban Institute — Justice Policy Center, Washington, DC, 2008.
513
[6] S. Kelty, R. Julian, Identifying the skills and attributes of good crime scene personnel,
Australas. Polic. 2 (2010) 40–41.
[7] J. Peterson, I. Sommers, D. Baskin, D. Johnson, The Role and Impact of Forensic
Evidence in the Criminal Justice Process, National Institute of Justice, 2010.
[8] H. McCulloch, Police use of forensic science, Police Research Series1996.
[9] B. Parker, The status of forensic science in the administration of criminal justice, Rev.
Jurid. Univ. P. R. 32 (1963) 405–420.
[10] J.L. Peterson, The utilization of criminalistics services by the police, U.S. Department
of Justice, Law Enforcement Assistance Administration, National Institute of Law
Enforcement and Criminal Justice1974.
[11] P.W. Greenwood, J.M. Chaiken, J. Petersilia, L. Prusoff, The criminal investigation pro-
cess, Observations and Analysis, vol. III, Rand Corporation, 1975.
[12] J.E. Eck, Solving Crimes: The Investigation of Burglary and Robbery, Police Executive
Research Forum, Washington, DC, 1983.
[13] D.A. Schroeder, M.D. White, Exploring the use DNA evidence in homicide investiga-
tions: implications for detective work and case clearance, Police Q. 12 (2009)
319–342.
[14] J. Burrows, R. Tarling, A. Mackie, H. Poole, B. Hodgons, Forensic science pathfinder
project: evaluating increased forensic activity in two English police forces, Home
Office Online Report 46/05, 2005.
[15] Home Office, DNA Expansion Programme 2000–2005: Reporting Achievement,
2005.
[16] R. Adderley, M. Townsley, J. Bond, Use of data mining techniques to model crime
scene investigator performance, Knowl.-Based Syst. 20 (2007) 170–176.
[17] M. Ramsay, The Effectiveness of the Forensic Science Service, Her Majesty's Statuary
Office, London, 1987.
[18] F. Horvath, R. Meesig, The criminal investigation process and the role of forensic ev-
idence: a review of empirical findings, J. Forensic Sci. 41 (1996) 963–969.
[19] N. Tilley, A. Ford, Forensic Science and Crime Investigation, Home Office, London,
1996.
[20] A. Barclay, Using forensic science in major crime inquiries, in: J. Fraser, R. Williams
(Eds.), Handbook of Forensic Science, Cullompton, UK 2009, pp. 337–358.
[21] R. Williams, Policing and forensic science, in: T. Newburn (Ed.), Handbook of
Policing, Willan, Cullompton 2007, pp. 760–793.
[22] J. Burrows, R. Tarling, Measuring the impact of forensic science in detecting burglary
and autocrime offences, Sci. Justice 44 (2004) 217–222.
[23] A. Ludwig, J. Fraser, R. Williams, Crime scene examiners and volume crime investi-
gations: an empirical study of perception and practice, Forensic Sci. Policy Manag.
Int. J. 3 (2012) 53–61.
[24] O. Ribaux, A. Baylon, C. Roux, O. Delémont, E. Lock, C. Zingg, P. Margot, Intelligence-
led crime scene processing. Part I: forensic intelligence, Forensic Sci. Int. 195 (2010)
10–16.
[25] J.G. Fraser, Not science…not support: forensic solutions to investigative problems,
Sci. Justice 40 (2000) 127–130.
[26] R. Williams, J. Weetman, Enacting forensics in homicide investigations, Polic. Soc. 23
(2013) 376–389.
[27] C. Wellford, J. Cronin, Clearing up homicide clearance rates, Natl. Inst. Justice J.
(2000) 2–7.
[28] L. Mucchielli, L'élucidation des homicides: de l'enchantement technologique à
l'analyse du travail des enquêteurs de police judiciaire, Déviance Soc. 30 (2006)
91–119.
[29] T.G. Keel, J.P. Jarvis, Y.E. Muirhead, An exploratory analysis of factors affecting homi-
cide investigations, Homicide Stud. 13 (2009) 50–68.
[30] D. Baskin, I. Sommers, The influence of forensic evidence on the case outcomes of
assault and robbery incidents, Crim. Justice Policy Rev. (2011) 1–25.
[31] N.P. Lovrich, T.C. Pratt, M.J. Gaffney, C.L. Johnson, C.H. Asplen, L.H. Hurst, T.M.
Schellberg, National Forensic DNA Study Report, Final Report, US Department of
Justice, Washington, DC, 2004.
[32] K.J. Strom, M.J. Hickman, Unanalyzed evidence in law-enforcement agencies,
Criminol. Public Policy 9 (2010) 381–404.
[33] O. Ribaux, A. Baylon, E. Lock, O. Delémont, C. Roux, C. Zingg, P. Margot, Intelligence-
led crime scene processing. Part II: intelligence and crime scene examination, Foren-
sic Sci. Int. 199 (2010) 63–71.
[34] J.L. Peterson, S. Mihajlovic, M. Gilliland, Forensic Evidence and the Police: The Effects
of Scientific Evidence on Criminal Investigations, US Government Printing Office,
Washington, DC, 1984.
[35] S.-A. Bradbury, A. Feist, The use of forensic science in volume crime investigations: a
review of the research literature, Home Office Online Report 43/05, 2005.
[36] B. Renard, P. Jeuniaux, Coûts et Pratiques autour des expertises ADN en matière
pénale, INCC, 2012.
[37] A. Marshall, The Principles of Economics, 8th edition Macmillan and Co., London,
1920.
[38] L. von Mises, Human Action. A Treatise on Economics, Ludwig von Mises Institute, 1998.
[39] D. Soergel, Indexing and retrieval performance: the logical evidence, J. Am. Soc. Inf.
Sci. 45 (1994) 589–599.
[40] D. Hazard, P. Margot, Forensic science culture, Encyclopedia of Criminology and
Criminal Justice, Springer 2014, pp. 1782–1795.
[41] A. Ludwig, J. Fraser, Effective use of forensic science in volume crime investigations:
identifying recurring themes in the literature, Sci. Justice 54 (2014) 81–88.
[42] O. Delémont, E. Lock, O. Ribaux, Forensic science and criminal investigation,
Encyclopedia of Criminology and Criminal Justice, Springer 2014, pp. 1754–1763.
[43] Association of Chief Police Officers, Using Forensic Science Effectively, ACPO, 1996.
[44] R.D. Julian, S.F. Kelty, C. Roux, P. Woodman, J. Robertson, A. Davey, R. Hayes, P.
Margot, A. Ross, H. Sibly, R. White, What is the value of forensic science? An over-
view of the effectiveness of forensic science in the Australian criminal justice system
project, Aust. J. Forensic Sci. 43 (2011) 217–229.
[45] O. Ribaux, S.J. Walsh, P. Margot, The contribution of forensic science to crime analy-
sis and investigation: forensic intelligence, Forensic Sci. Int. 156 (2006) 171–181.
[46] N. Tilley, M. Townsley, Forensic science in UK policing: strategies, tactics and effec-
tiveness, in: J. Fraser (Ed.), Handbook of Forensic Science, Willan, Cullompton 2009,
pp. 359–379.