Torres, A.N., Boccaccini, M.T., and Miller, H.A. (2006). Perceptions of the validity and utility of criminal profiling among forensic psychologists and psychiatrists, American Psychological Association, 37 (1), 51-58.
Study purpose, research topic, and research questions. This research explores the perceptions of forensic psychologists and psychiatrists in regard to the utility and validity of criminal profiling. It is important to first establish a clear definition of the key term used in this research: Criminal profiling. The definition of criminal profiling in this research is not the same as in the vernacular. The authors discriminate forensic criminal profiling as the use of "behavioral evidence left at a crime scene to make inferences about the offender, including inferences about personality characteristics and psychopathology" (Torres, et al., 2006, p. 51). From the literature (Davis & Follette, 2002), the authors overly the simplest of definitions: "…profiling is simply the postdiction of behavior; an action has taken place that allows investigators to make inferences about the person responsible" (Torres, et al., 2006, p. 51). Older television viewers of Columbo and contemporary viewers of the many variants of CSI will find this a familiar cognitive path. However, the sleuthing function of forensic criminal profiling is more limited in real-life investigations, according to Douglas and Olshaker (1995), with the main goal refined to "narrow the scope of a suspect pool rather than to identify a single guilty criminal" (Torres, et al., 2006, p. 51).
The researchers argue that, despite misgivings about its utility and scientific validity, the use of psychological profiling in criminal investigations is on the rise. The primary purpose of this research is "use a survey approach to explore perceptions of the validity and utility of criminal profiling among forensic psychologists and psychologists" (Torres et al., 2006, p. 53). The authors pose several related research questions that focus on the proportion of professionals who use a criminal profiling approach, differences in the professionals' perceptions about its use, and whether a change in the name attributed to the procedure has impacted perceptions about validity and overall acceptance of the approach. Criminal profiling is referred to by many different names that are dependent, in part, on the professional field in which they are utilized, and also in order to increase acceptance of the process in criminal proceedings. For instance, the authors refer to the term criminal investigative analysis that is currently used by the FBI. The literature provides examples of how the terms used to refer to criminal profiling have influenced the acceptability of its use by the courts. Exploration of the impact of this change in terminology is an integral part of this study.
Research design and approach. The research employed quantitative research methods using a survey approach. The survey data were quantitative and were statistically analyzed to generate findings which were shown in tabular form in the article. Data was reduced to composite scores and was treated quantitatively. No open ended questions were used in the survey and narrative data was not elicited from the respondents.
Sampling. The sampling methodology was purposive in order to ensure the sample population reflected the mental health and law professional associations whose members were likely to be in a position to utilize criminal profiling. Specifically, the researchers cross-checked association membership lists to ensure that professionals who received surveys were psychologists and not attorneys, graduate students, or other affiliate members. Some professional organizations permitted the researchers to email their member directly, others required that the presidents of those organizations send an introductory email inviting their participation in the survey. Emails soliciting participation in the survey were sent to 147 AAFP members, 175 AP-LS members, 840 AAPL members, 475 members of the Police and Public Safety section of the APA Division, and 1,637 forensic mental health professionals. Of those who received the email solicitations, only 9.9% (N=161) completed surveys. Low return rates for surveys provided on Websites or via the Internet are characteristically lower than the return rates for telephone and mail surveys (Schonlan, Fricker, & Elliot, 2002). Further, a meta-analysis of the research regarding response rates for surveys provided by academic and professional groups tend to be lower than for other groups (Cook, Heath, & Thompson, 2000). The limitations of the study that are related to the low rate of return for completed surveys are thoroughly addressed by the researchers, who caution readers to interpret the survey results with the limitations clearly in mind.
Survey instrumentation. One of the primary research objectives was to assess the differences between perceptions related to utility, validity, and acceptance of criminal profiling methods, based on whether the profiling was described as criminal investigative analysis or profiling. To aid the investigation into these potential differences, two nearly identical versions of the survey were created, with the only distinguishing feature the use of profiling in one set of surveys and the use of criminal investigative analysis in the other set. The survey instrument consisted of three sets of five questions, which were categorized according to the emphasis of the question items. Question items 1 -- 5, for which respondents were asked to mark either yes or no, were related to the personal attitudes, professional knowledge, training, and experiences related to criminal investigative analysis or profiling. Question items 6 -- 10, for which respondents were also asked to mark either yes or no, focused on the scientific and professional status of criminal profiling or criminal investigative analysis. The final five question items 11-15, for which respondents were asked to mark either correct or incorrect, assessed the respondents knowledge of the standards courts use to admit expert testimony into court . The responses to the question items in this last section were totaled to form a composite score of admissibility knowledge. These composite scores were found to have acceptable levels of internal consistency (Kuder-Richardson 20 = .72).
Research findings. One significant difference was found between psychologists and psychiatrists on the main survey questions related to their training, experience, knowledge, and attitudes about profiling. This single effect indicated that psychiatrists were somewhat more likely to describe profiling as a useful law enforcement tool (95.4%) than were psychologists (84.9%), with mean and significance parameters for this item as follows: X2 (1, N = 172) = 4.12, p < .05, ? = .17. Although the difference is statistically significant, it is important to note that the majority of respondents believed that profiling was a useful law enforcement tool, regardless of their professional orientation. In fact, the similarities in response patterns for both groups led the researchers to report findings the entire sample of forensic mental health professionals as a group, rather than segregating psychologists from psychiatrists.
A sizable minority of forensic mental health professionals indicated that they had knowledge of and training in profiling and criminal investigative analysis, but only 10% had ever actually generated a profile and only 10% had ever actually testified in court about their use or opinions of profiling techniques. Fewer than half of the respondents viewed profiling or criminal investigative analysis as reliable, valid, or sufficiently scientific to be admitted into court. And perceptions were lowest for those respondents whose surveys used the term profiling. All differences between profiling and criminal investigative analysis failed to reach clinical or statistical significance (p >.05, phi coefficients (?) < .06). The researchers concluded that profiling by another name is viewed more favorably. Nearly all forensic mental health professionals stated that empirical research should be carried out to investigate profiling and criminal investigative analysis.
Correlation was found between levels of knowledge of admissibility standards and beliefs related to the scientific merit of profiling and criminal investigative analysis for psychologists, but not for psychiatrists. The operational definition of scientific merit was generated by totaling the yes responses to survey questions about the reliability, validity, and scientific support for profiling to achieve a composite scientific merit score. Scores on this composite could range from 1 to 3, with the higher numerical score indicating positive attitudes about the scientific merit of profiling. Psychologists with higher levels of knowledge about admissibility viewed profiling and criminal investigative analysis less favorable with regard to scientific merit (r8 = -.03, p > .05).
Both profiling and criminal investigative analysis were viewed by more than half of all respondents as lacking adequate reliability, validity, and empirical scientific support. Interestingly, few forensic mental health professionals responding to the survey had ever conducted a profile or criminal investigative analysis, yet a large minority of these respondents considered themselves knowledgeable about the techniques. The researchers posit several scenarios for this phenomenon, including the notion that most forensic mental health professionals appear to be taking a wait-and-see approach. Further, the data from this study indicate that mental health practitioners have not moved away from opinions held over ten years ago when Bartol (1996) found that 70% of police psychologists "seriously questioned" the validity of profiling (p.79).
The researchers recommend that practitioners in the forensic mental health field consider carefully the influence that terminology has on the general and professional acceptance…