PSYCHOLOGICAL PROFILING:
TENUOUS VALIDITY AS EVIDENCE FOR CONVICTION
IN U.S COURTROOMS
Emily Charrier
Robert Miller
Quick Yeates-Burghart
Psychology and Law
University of Oregon
2
Introduction ......................................................................................................................... 3
Argument ............................................................................................................................ 4
Limitations of Small Data Pool ...................................................................................... 4
Non-standardized Profiling Methodology ...................................................................... 7
Diversity of Background Fields lead to lack of Communication Between Profilers ...... 8
Consequence of Evidence ............................................................................................. 11
Opinion on Ultimate Issue ............................................................................................ 12
Constitutionality & Probable Cause ............................................................................. 12
Conclusion ........................................................................................................................ 13
References ......................................................................................................................... 15
3
INTRODUCTION
Psychological profiling has gained attention as a double-edged sword in U.S
courtrooms. Popular media has cast it as a solid, predicting force for the behavioral
patterns and actions of violent criminals. The scientific community and subsequent
literature contest the claims of objective, scientific validity within a courtroom setting.
Whether profiling should be allowed into a courtroom as a piece of evidence against a
defendant is ultimately decided by a judge on a case-by-case basis. In this case, the
prosecution attempted to submit a serial killer’s personality profile as evidence to the
court, claiming that the defendant matched the characteristics thereof. The court ruled in
favor of the prosecution’s request, and the defendant was convicted. This brief will
contest the correctness of the courts decision to allow the profile to be admitted as
evidence.
The primary support for disallowing psychological profiling into a case comes
from the Federal Rules of Evidence. Rule 702 as amended in December 2001 states that
the only time expert testimony is admissible in court is when, “(1) the testimony is based
upon sufficient facts or data, (2) the testimony is the product of reliable principles and
methods, and (3) the witness has applied the principles and methods reliably to the facts
of the case.” Psychological profiling is a relatively young discipline that uses nonstandardized methodologies and lacks the scientific reliability that Rule 702 requires.
There has not been sufficient testing on the practice to determine if it is effective or
accurate enough for courtroom applications (Mauro, 2005).
In addition, this brief will show that there are two distinctly different methods of
profiling and that they are not equivalent. Depending on the methodology utilized to
4
compile a profile, it cannot be corroborated that the expert witness “applied the principles
and methods reliably to the facts of the case”. Rule 403 states that evidence can be
thrown out if the danger of introducing bias to the jury outweighs the value of the
evidence. Claiming that the defendant fits a criminal profile has an immediate potential to
influence the jury’s perception of the case. Due to doubts about the scientific reliability of
profiling and the danger of introducing prejudice to the jury, stating that the defendant
fits a criminal profile should not be admissible in court as evidence.
ARGUMENT
Limitations of Small Data Pool
Profiling was not a frequently used tool until the 1970’s when the FBI developed
a Behavioral Sciences Unit. FBI investigators Robert Ressler and John Douglas helped
to create one of the current standards of profiling used modernly. Throughout the 70’s
and 80’s Ressler and Douglas interviewed 36 serial killers that had been caught and
imprisoned (Parker, 2002). Among this group were several “celebrity” killers such as
Charles Manson and David “Son of Sam” Berkowitz. Based on these interviews a
number of statistics were drawn up of what a “typical” serial killer’s psychological
profile would contain (Glazer, 2003). Today, FBI agents continue to interview serial
killers to add to this profile.
Most commonly, a profiler compares personality traits and observed behaviors of
an individual to the recorded behaviors of a small, under-represented group of offenders.
Statistical information derived from a small sample results in large confidence intervals
and an inaccurate pool for the profiler to draw conclusions from. Many studies in this
5
vein caution that there are fundamental problems with predicting behavior based on a
limited, known group (Turvey, 1995). Ideally, researchers try to achieve large samples
that are representative of the mean characteristics that describe a population (Gravetter &
Walneau, 2002).
Due to the scarcity of serial killers, estimated to commit less than one percent of
all homicides, it is apparent that there is still a good deal left to learn (Glazer, 2003).
There are many theories about the life events that create a serial killer, ranging from
psychological problems to abusive childhoods. It is generally accepted that some
psychosocial background traits have been linked to an increased likelihood that violent
offenders will continue to commit violent crimes in the future (Valenzuela, Villanueva,
Aguilera & Luna, 1991). However, it is important to note that while results may be
indicative of future violence, they do not suggest murder (Pinta, Bessimer, Smeltzer &
Fine, 1990). Ultimately, there is no definitive answer as to what drives a person to
become a serial murderer. We do not understand serial killers and therefore cannot make
conclusions through profiles about whom they are. It is tenuous to infer any concrete
behaviors or personality patterns from this limited data (Glazer, 2003).
Yet, there is some advantage in having a small, non-representative sample to use
as the standard. Small samples generally have less variability than whole populations. In
some instances a small sample can also have a small variability if the subjects are
homogeneous and different from the population. One argument for psychological
profiling is that serial killers are similar to one another behaviorally, but different from
the general population. Following this, any statistically supportable facts that can be
obtained from the pool may be incorporated into or used in comparison to the profile of
6
the suspect. On the other hand, by drawing conclusions from this incomplete body of
information, the profiler must make some inferences. Primarily, that the behavior and
characteristics of the individual in question can be generalized and predicted from
statistical analyses of the same traits within a population of rare criminals (Turvey, 1998).
This is partially supported by the literature, but only for short-term predictions (Hall,
Catlin, Boissevain & Westgate, 1984). In the recent case of the Washington D.C. snipers,
profilers fell flat when the young, solitary white male working alone turned out to be a
“father-son” pair of black men working together (Parker, 2002).
Additionally, the profiler must assume that an individual’s behavior and
motivation are static, predictable features rather than novel or spontaneous developments
within a unique being. There is accepted recognition in the literature that criminal
motives and behavior patterns are diverse, rather than static (Cornell, Benedek &
Benedek, 1987). Moreover, many questions have been raised about the reliability of selfreported survey and interview information submitted by criminals (Godwin, 1998). There
is no way of knowing what aspects of an interview are truthful. Their reliability can be
significantly lessened in the face of recorded inconsistencies. For example, Ted Bundy
never admitted the actual number of killings he committed. His claims ranged from 3 to
300 and nothing was ever proven (Glazer, 2003). Furthermore, when asked why he
committed the murders, he couldn’t coherently explain his thoughts or reasons (Glazer,
2003). If someone so familiar with the act of murder cannot articulate his own reasoning
on the subject, how can anyone who never felt the urge to kill do so? (Glazer, 2003).
Finally, it is important to note that these interviews are based only on killers that have
been caught. They exclude an entire demographic of serial killers that remain
7
undiscovered or free in the population. For example, Jack-the-Ripper authored a series of
letters to the police, but was never apprehended. Dr. Bond’s profile was based both on
these letters and forensic evidence; however, the accuracy of his profile will never be
known.
Non-standardized Profiling Methodology
Psychological profiling is a complex topic that cannot be argued for or against
without additional clarification of the methodologies utilized. There are two generally
recognized methods to compile a profile, deductively or inductively. However, they are
dichotomous and generally unequal in validity. Deductive profiling is rarely used in the
courtroom and is usually not the focus of scientific critique, nor the focus of this
argument. The deductive process concentrates on cases of individual offenders by
compiling unique psychological profiles across a relatively long period of time. It makes
use of physical evidence to reconstruct the signature behaviors and motivations of violent
criminals (Turvey, 1998). Inductive profiling is based on statistical figures drawn from
known populations of criminals, intuitive or anecdotal guesses on the part of the profiler
and public data sources (Holmes, 1989). This technique is simple to use, requires little
experience and is a fast way to build a general profile (Turvey, 1998). It is the more
popularly utilized, glamorized and criticized method. Arguments against the inductive
method rally around three important issues. First, it uses statistical information from a
small, known population of criminals to infer generalized information about individuals.
Second, the data collected and used to create a profile comes from a totally known
source. It is incapable of taking criminals outside this recognized set into account. This
8
leads to a one-sided profile that comes from an unrepresentative sample. Third, the forced
ambiguity of a profile compiled by the inductive method will naturally carry a percentage
of inaccurate details. The descriptions included in profiling are vague and can be
interpreted to encompass a wide variety of people. Under Rule 403 of the Federal Rules
of Evidence, “evidence may be excluded if its probative value is substantially outweighed
by the danger of unfair prejudice.” An over-inclusive profile may serve to confuse the
jury’s impressions of a defendant by introducing vague information with limited
relevance to the case (Bailey, 2001). These details and exuberant profilers have been
known to complicate investigations with false positives (Alison, Bennell, Mokros &
Ormerod, 2002).
The differences between inductive and deductive profiling methodology create a
breach in the continuity of the underlying theory, making it difficult to accept using these
processes in a courtroom setting. Both sides claim that their research is valid, yet peck at
one another in the literature over details (Baeza & Mcgrath, 2000). Such conflict within
the field implies that the principles and methods are neither well understood nor cohesive.
How could the testimony from an expert in the field be trusted when the literature itself
cannot decide who the experts are? The debated lack of accuracy makes profiles
inadmissible in court under rule 702, and should not be entered in this case.
Diversity of Background Fields lead to lack of Communication Between
Profilers
The discipline of profiling has been developed from bits and pieces of many
scientific fields. It draws individuals from the ranks of forensic psychology and
9
psychiatry, criminology, the FBI, state and local law enforcement and others (Turvey,
1998). Due to this, profilers as a group tend to be varied in their methodologies,
interpretation styles, compiling procedures and personal experiences (Jackson &
Bekerian, 1997). This variation is criticized as one of its largest deficiencies. For one, the
professionals typically involved in profiling do not always have well-rounded training in
forensic psychology, psychopathology, investigative procedure or specific experience
with investigating serial killers (Turvey, 1998). In addition to being varied as a group,
each profiler uniquely applies methodology and techniques. Profiling is not a uniform or
standardized discipline as such it is difficult to justify its use in courtrooms. Rule 702
states that the testimony of the expert must be “the product of reliable principles and
methods”. If a profiler has no practical field experience with these fundamental aspects of
criminal psychology and investigation, how are their conclusions valid? Furthermore,
how are they able to testify as expert witnesses on a subject they not an expert on?
To date there is no agreed upon definition or typographical model for violent
serial criminals. A major hurdle for both the fields of mental health and law enforcement
is finding common experience enough to standardize a definition and classification
process (Turvey, 1995). There is little overlap between the disciplines of crime scene
investigation and psychology, psychopathology and human behavioral study. This lack of
cohesiveness leads to a confusion of dialect definitions between law enforcement
agencies and forensic social scientists. Profiles compiled on the same subject by different
profilers will often be radically different because of these dialectical differences (Jackson
& Bekerian, 1997).
10
The field of profiling has been criticized for the differences in opinions between
professional profilers. In a recent study by Kocsis (2003) participating profilers were
accurate as a group, but rather inconsistent as individuals (of 47 invited profilers, only 11
chose to participate in the study). This large variance in expert opinions indicates a lack
of theoretical cohesiveness and scientific reliability in the processes involved in
developing criminal profiles. Inconsistency of this magnitude brings Rule 702 into sharp
relief. Such uncertainty is problematic because it raises the question of whether the
profile compared to the defendant was reliably compiled.
There is intense debate within the fields that comprise the discipline of profiling
in regard to its legitimacy as a scientific instrument (Glazer, 2003). The principles,
methodology and individuals that make up this discipline are in disarray. Lack of
communication within and between professional groups involved is evident. Due to the
lack of agreement over this practise among the professionals in the field, including a
profile in court is against the ruling set forth by US vs. Frye (1923).
State vs. Fortin, State vs. Haynes and State vs. Anthoney are three cases where
judges have refused to allow criminal profilers to testify, or allowed criminal profiles to
be used as character evidence. “Courts refusal to admit into evidence criminal
psychological profiles is based on the grounds that they constitute impressible character
testimony, improper subject for scientific or expert testimony, and unduly prejudicial
evidence” (Ingram, 1998). The court decisions in these cases should be upheld and
applied until science can prove the reliability of psychological profiles. As such the
profile in this case should be inadmissible in evidence as well.
11
Consequence of Evidence
It is not known under which rules of evidence the court allowed either the profile
or the profiler’s opinion. As previously discussed, the three conditions leading to the
inclusion of evidence with Rule 702 are that there is sufficient data, reliable methods or
principles, and that it is applied correctly. Under Rule 702 it is apparent that the court has
erred because in order to include testimony all three conditions must be met. The deficit
of any of the conditions would exclude the profile from use as evidence. As shown, there
is a very probable deficit in all conditions due to the poor state of the art of psychological
profiling.
If the court included the evidence under Rule 703, which states that, “If of a type
reasonably relied upon by experts in the particular field in forming opinions or inferences
upon the subject, the facts or data need not be admissible in evidence in order for the
opinion or inference to be admitted,” then the trial court erred as well. This brief shows
that the dialectical and methodological differences would not permit room for facts or
data that could be reasonably relied upon to form opinions by a consensus of experts
because they differ so greatly. Even under the assumption that the facts and data fit this
condition, the inclusion of the profile in trial court is unjustifiable as “Facts or data that
are otherwise inadmissible shall not be disclosed to the jury by the proponent of the
opinion or inference unless the court determines that their probative value in assisting the
jury to evaluate the expert's opinion substantially outweighs their prejudicial effect.”
(Rules of Evidence, 703). Profiles consistently use ambiguous language and the Barnum
effect is likely to prejudice the jury, as it has prejudiced police officers and other subjects
12
in Kocsis’ studies (2004; I, II, III). The assumption that the facts and data could be
reasonably be relied upon is not relevant owing to Rule of Evidence 704(b).
Opinion on Ultimate Issue
Rule of Evidence 704(b) specifies that “No expert witness testifying with respect
to the mental state or condition of a defendant in a criminal case may state an opinion or
inference as to whether the defendant did or did not have the mental state or condition
constituting an element of the crime charged or of a defense thereto. Such ultimate issues
are matters for the trier of fact alone.” A psychological profile is a collection of
inferences and opinions regarding the mental state or condition of the defendant. The
prosecution introduced evidence to show that the defendant fits the profile of a serial
killer. Because the evidence and testimony infer that the defendant’s mental condition –
allegedly that of a serial killer – constitutes an element of the crime charged with, it is
inadmissible in court. Under rule 704(b) the profiler is not permitted to provide testimony
on the requisite mens rea of the defendant.
Constitutionality & Probable Cause
The United States legal system favors “the right of the people to be secure in their
own persons… against unreasonable searches and seizures… but upon probable cause ”
(Bill of Rights; Amendment IV). Profiles at the time of this writing successfully identify
offenders only 17% of the time (Kocsis, 2003). A 17% success rate is not enough to make
a case for probable cause. It is not constitutionally sound to convict a person solely on the
13
grounds that they fit a description. U.S. v. Lopez (1971) questioned the wisdom of
detaining potential violators on the grounds that they might commit crimes in the future:
Furthermore, if we optimistically predict that in a pool of 1000 defendants, 17%
were convicted and 90% of those are actually guilty, then we have set 83 guilty people
free and convicted 17 innocent people.
Guilty
Innocent
Sentenced
Not Sentenced
(.17)(.9) x 1000 = 153
(.83)(.1) x 1000 = 83
236
(.17)(.1) x 1000 = 17 (.83)(.9) x 1000 = 747
754
(.17) x 1000 = 170
(.83) x 1000 = 830 1000
This means that profiling is violating the constitution more often than it is offering justice
when using unrealistically conservative figures. We must still take into account
inaccuracies such as profile ambiguity, profiler bias and attempts to execute a speedy trial
by the legal system – reducing the amount of forensic evidence – and thus rendering our
optimistic numbers inadequate. The probabilities of convicting innocent people and
permitting criminals to go free would outweigh the probabilities of convicting the actual
criminal.
CONCLUSION
The laws of evidence regarding expert testimony were set forth by several court
cases. In the case of US v Frye (1923), the U.S. Supreme Court ruled that a science must
14
be generally accepted in its field. In Dyas vs. US (1977) the courts ruled that the “state of
art must be sufficiently advanced”. This second ruling means that the practice must have
a solid scientific basis. Profiling fits into neither of these categories. As shown through
the posed arguments, the experts in the field of psychology, forensics, and law
enforcement cannot agree on a standardized methodology, who is an “expert”, whether or
not there is validity in inductive profiling or what direction the field should move in. The
scientific basis for profiling is questionable. Dichotomous methodology makes it difficult
to compare profiles scientifically. Those compiled using the inductive method are often
subjectively biased. Also, there is not enough known about serial killers to make accurate
assumptions regarding their motives and behavioral characteristics.
Profiling may be a potential way to limit suspects in an on-going investigation.
However, it is not yet state-of-the-art and is inadmissible in the courtroom as testimony
on an ultimate issue. At this point in time, more research needs to be done on
psychological profiling before it can legitimately be used as evidence in a trial. The trial
court has erred in its decision to allow the testimony and evidence of a profiler and
should reverse its decision.
15
REFERENCES
Alison, L., Bennell, C., Mokros, A.; Ormerod, D. (2002). The Personality Paradox in
Offender Profiling: A Theoretical Review of the Processes Involved in
Deriving Background Characteristics From Crime Scene Actions. Psychology,
Public Policy & Law. Vol. 8(1).
Alison, L., Smith, M., Morgan, K. “Interpreting the Accuracy of Offender Profiles”
University of Liverpool. 26 Mar 2001.
Baeza, J., Mcgrath, M. (2000). Review of Reliability, Validity and Utility of Extant Serial
Murder Classifications. Journal of Behavioral Profiling. Vol. 1(2).
Bailey, K. A. (2001). Legal implications of profiling students for violence. Psychology in
the Schools. Vol. 38.
Bardsley, M. “All For Love” 22 Jan 2005. <www.crimelibrary.com>.
Bekerian, D.A. and Jackson, J.L. (1997). ‘Critical Issues in Offender Profiling’ in
Offender Profiling: Theory, Research and Practice. Eds. Jackson, J.L. and
Bekerian, D.A. John Wiley & Sons, New York.
Brown, P. “The Real Serial Killer” 16 Jan 2005. <www.crimelibrary.com>.
Campbell, T.W. (1999). Challenging the Evidentiary Reliability of DSM-IV.
American Journal of Forensic Psychology. Vol. 17(1).
16
Cornell, D.G., Benedek, E.P., Benedek, D.M. (1987). Characteristics of Adolescents
Charged with Homicide: Review of 72 Cases. Behavioral Sciences & the Law.
Vol. 5(1).
Dyas v. U.S. Supreme Court. (1977).
Fischer, M. (1997). The Psychologist as a “Hired Gun”. American Journal of Forensic
Psychology, Vol. 15(2).
Freedman, D. (2001). False Prediction of Future Dangerousness: Error Rates and
Psychopathy Checklist-Revised. Journal of the American Academy of Psychiatry
and Law. Vol. 29.
Glazer, S. (2003). “Do We Know Enough To Catch Them?” CQ Researcher Online
13.38 .23 January 2005 <http://0library.cqpress.com.janus.uoregon.edu
:80/cqresearcher/cqresrre2003103100>. Document ID: cqresrre2003103100.
Godwin, G. M. (1998). Reliability, Validity and Utility of Extant Serial
Murder Classification Models. The Criminologist, Vol. 22(4).
Gravetter, F.J., Wallnau, L.B., (2002). Essentials of Statistics for the Behavioral
Sciences. 4th Ed. Thomson Learning, Australia.
Federal Rules of Evidence. 403; 702; 703; 704(b)
Hall, H., Catlin, E., Boisseuain, A., Westgate, J. (1984). Dangerous Myths About
Predicting Dangerousness. American Journal of Forensic Psychology. Vol.2.
Hare, R.D. (1998) The Hare PCL-R: Some Issues Concerning Its Use and Misuse.
Legal and Criminological Psychology. Vol. 3.
17
Holmes, R. M. (1989). Profiling Violent Crimes: An Investigative tool. Sage
Publications, New York.
Ingram, S. (1998). “If the Profile Fits: Admitting Criminal Psychological Profiles into
Evidence in Criminal Trials”. Journal of Urban and Contemporary Law. Vol.
54 (239).
Johnson, C., and Scott, B. (1976). “Eyewitness testimony and suspect identification
as a function of arousal, sex of witness, and scheduling of interrogation.” Paper
presented at the annual meeting of the American Psychological Association,
Washington, D.
Kocsis, R. N. (2003). “Criminal psychological profiling: Validities and abilities?”
International Journal of Offender Therapy and Comparative Criminology. 47(2).
Kocsis, R. N., & Hayes, A. F. (2004). “Believing is seeing? Investigating the perceived
accuracy of criminal psychological profiles.” International Journal of Offender
Therapy and Comparative Criminology. 48(2).
Kocsis, R. N., & Heller, G. Z. (2004). “Believing is seeing II: Beliefs and perceptions of
criminal psychological profiles.” International Journal of Offender Therapy and
Comparative Criminology 48(3).
Kocsis R.N., Middledorp J. (2004). “Believing is seeing III: perceptions of content in
criminal psychological profiles”. International Journal of Offender Therapy and
Comparative Criminology. 2004 Aug; 48(4).
Parker, L. “Profiling: The Art of the Educated Hunch” USA Today 11 Jan 2002.
18
Pinta, E.R. Bessimer, P., Smeltzer, D., & Fine, R. (1990). Background Variables in
Murderers, Violent and Nonviolent Felons. American Journal of Forensic
Psychology, Vol. 8.
Pinizzotto, A. (1984). Forensic Psychology: Criminal Personality Profiling. Journal of
Police Science and Administration. Vol. 12(3).
Ramsland, K. “The Profile Evaluated” 16 Jan 2005. <www.crimelibrary.com>.
Turvey, B.E. (2003). Forensic Frauds: A Study of 42 Cases. Journal of Behavioral
Profiling. Vol. 4(1).
Turvey, B.E. (1998) Deductive Criminal Profiling: Comparing Applied Methodologies
Between Inductive and Deductive Criminal Profiling Techniques.
Knowledge Solutions Library. <http://www.corpusdelicti.com/Profiling_Law.html>
Turvey, B.E. (1995). The Impressions of a Man: An Objective Forensic Guideline
to Profiling Violent Serial Sex Offenders. Knowledge Solutions Library.
<http://www.corpud-delicti.com/impress.html>
U.S. Constitution. Bill of Rights.
United States. Cong. House. Committee on the Judiciary. The Federal Rules of
Evidence. 107th Cong., 1st Sess. Washington: 2001.
U.S. v. Frye, Supreme Court (1923).
U.S. v. Lopez, Supreme Court (1971).
19
Valenzuela, A., Villanueva, E., Aguilera, J., and Luna, A. (1991). Importance of
Criminal Behavior Related Variables in Criminal Prognosis. American Journal of
Forensic Psychology. Vol. 9(2).
Wyda, J. and Black, B. (1989). Psychiatric Predictions and the Death Penalty: An
Unconstitutional Sword for the Prosecution but a Constitutional Shield for the
Defense. Behavioral Sciences & the Law. Vol. 7(4).