Thursday, January 30, 2014

Research roundup

The articles are flooding in at an alarming rate, threatening to bury me under yet another avalanche. Before I am completely submerged, let me share brief synopses of a few of the more informative ones that I have gotten around to reading.


Assessor bias in high-stakes testing: The case of children’s IQ


I’ve blogged quite a bit about bias in forensic assessment, reporting on problems with such widely used tests as the Psychopathy Checklist and the Static-99R. As I’ve reported, some of the bias can be chalked up to adversarial allegiance, or which side the evaluator is working for, whereas some may be due to personality differences among evaluators. Now, researchers are extending this research into other realms -- with alarming findings.


In a study of intelligence testing among several thousand children at 448 schools, the researchers found significant and nontrivial variations in test scoring that had nothing to do with children’s actual intelligence differences. The findings, reported in the journal Psychological Assessment, are especially curious because scoring of the test in question, the Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV), seems relatively straightforward and objective (at least as compared to inherently subjective tests like the Psychopathy Checklist, for example).


The article is:

  • Whose IQ Is It? Assessor Bias Variance in High-Stakes Psychological Assessment.  McDermott, Paul A.; Watkins, Marley W.; Rhoad, Anna M. Psychological Assessment, Published online on Nov 4 , 2013. To request a copy from the first author, click HERE.





Beware pseudo-precision in expert opinions


I’ve never forgotten a video I saw a long time ago, in which the filmmakers drove up to random strangers and asked for directions to a nearby landmark. Some of the good samaritans gave enthusiastic instructions that were completely wrong, while other people gave correct directions but in a more tentative fashion. The trouble is, the more confident someone appears, the more we judge them as knowing what they are talking about.  


One way we gauge a presenter’s confidence, in turn, is by their level of precision. In a new study, researchers found that participants were more likely to rely on advice given by people who provided more precise information. For example, they were more likely to trust someone who said that the Mississippi River was 3,992 miles long, rather than 4,000 miles long.


What this means in the forensic realm is that we should not make claims of false precision, when our evidence base is weak. For example, we should not claim to know that someone has a 44 percent chance of violent reoffense within three years. Such misleading claims-making lends an aura of confidence and expertise that is not warranted.


The article is:




Ethics and the DSM-5


Speaking of avalanches, the volume of critical response to the DSM-5 is lessening now that the tome has been on the bookshelves for eight months. Trying to keep my finger on the pulse because of my training activities on the manual’s forensic implications, I found an interesting summary of the ethical dilemmas of the latest trends in psychiatric diagnosis.


The author, Jennifer Blumenthal-Barby, is an ethics professor at Baylor College of Medicine’s Center for Medical Ethics and Health Policy. In her critique, published in the Journal of Medical Ethics, she focuses on consequence-based concerns about the dramatic expansion of psychiatric diagnoses in the latest edition of the American Psychiatric Association’s influential manual. Concerns include:


  • False positives, or over-diagnosis, in clinical (and I would add forensic) practice
  • Risks associated with pharmacological treatments of new conditions
  • Neglect of larger structural issues and reduction of individual responsibility through medicalization
  • Discrediting of psychiatry through the trivialization of mental disorders
  • Efforts to eradicate conditions that are valuable or even desirable


Although her discussion is fairly general, she does mention a few of the proposed diagnostic changes of forensic relevance that I’ve blogged about. These include the proposed hypersexual disorder and a proposal to eliminate the age qualifier (of 18 and above) for antisocial personality disorder, to make it consistent with all of the other personality disorders.


It’s a good, brief overview suitable for assignment to students and professionals alike.


The article is: 
  • Psychiatry’s new manual (DSM-5): ethical and conceptual dimensions. Journal of Medical Ethics. Published online on 10 Dec. 2013. To request a copy, click HERE.




Dual relationships: Are they all bad?


We’ve all seen the memo: Dual relationships are to be avoided.


But is that always true?


Not according to ethics instructor Ofer Zur.


Multiple relationships are situations in which a mental health professional has a professional role with a client and another role with a person closely related to the client. In a new overview, Zur asserts that, not only are some multiple relationships ethical, they may be unavoidable, desirable, or even -- in some cases -- mandated.


In delineating the ethics and legality of 26 different types of multiple relationships, Zur stresses that in forensic settings, most multiple relationships should be avoided.


The article, Not All Multiple Relationships Are Created Equal: Mapping the Maze of 26 Types of Multiple Relationships, is another good teaching tool, and is freely available online at Zur’s continuing education website.

By the way, if you are in California and are looking for more ethics training, Zur and two of my former colleagues from the state psychological association’s Ethics Committee -- Michael Donner, PhD and Pamela Harmell, PhD -- are co-presenting at an interactive ethics session at the upcoming California Psychological Association convention. The convention runs April 9-13 in Monterey, and the ethics conversation -- “Ethics are not Rules: Psych in the Real World” -- is on Saturday, April 12.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.

 
Real Time Web Analytics