Timothy M. Tippins
Timothy M. Tippins (NYLJ/Rick Kopstein)

Attorneys and judges involved in custody litigation frequently rely upon the reports and testimony of forensic evaluators. Unfortunately, not all evaluations are reliable. Any number of dynamics can diminish or destroy the reliability of forensic work-product. One of the most pervasive and pernicious of these is bias. To be informed consumers of forensic work-product, lawyers and judges need to understand the insidious nature of bias in its many forms and must know how to identify it in reports and testimony. Additionally, attorneys need to develop trial strategies that will expose bias, which is often hidden beneath the surface of the report, so that the trier of fact can see it plainly and appreciate its insidious impact on forensic reliability.

This article will address the nature of bias and discuss why it is so prevalent in the custody setting. Ensuing articles will deal with the challenges of identifying it in forensic evaluations and exposing it in the courtroom.

Bias as a Mode of Impeachment

Attacking the credibility of a witness by showing that he or she is biased is a well-entrenched method of impeachment. As Wigmore noted, “a partiality of mind is therefore always relevant as discrediting the witness and affecting the weight of his testimony.” Wigmore on Evidence, Chadbourne Revision, Section 940 (Aspen 1970). Bias is so central to credibility that it is deemed “non-collateral,” meaning that the cross-examiner is not bound by the witness’s denial of the impeaching question and may introduce extrinsic evidence to prove the bias. Id., Section 948.

Decisional law most often speaks of bias in terms of a witness’s relationship to one of the parties or to some interest of the witness in the outcome of the case. Coleman v. New York City Transit Authority, 37 N.Y.2d 137 (1975); see also People v. Webster, 139 N.Y. 73 (1893). In the context of expert witnesses, case law has held that the fact and extent of the expert’s compensation to be relevant to bias. Zimmer v. Third Avenue R.R. Co., 36 App. Div. 273 (2d Dept. 1899)

But these are merely exemplars of bias, not a definition. As a delineation, they are unduly narrow. Attorneys and judges need to understand bias in its broader sense.

Bias More Broadly Understood

The American Psychological Association [APA] defines bias as “an inclination, tendency, or preference.” APA Dictionary of Psychology (American Psychological Association, 2007, p. 116. Jeffrey P. Wittmann, author of “Evaluating Evaluations: An Attorney’s Handbook for Analyzing Child Custody Reports” (MatLaw Systems 2013), provides a meaningful definition:

Bias in the broadest sense refers to an emotional or cognitive inclination that interferes with an unprejudiced consideration of the data that has been gathered.

Id. at p.160.

David A. Martindale, a prominent and prolific forensic psychologist, expounds upon this further:

Bias … includes prejudgments but also refers to attitudes, beliefs, or judgments that may be spontaneously triggered by events that occur during the evaluative process or that may be attributable to the methodology employed by the evaluator.

Martindale, D.A., “Bias and Prejudice in Custody Evaluations,” The Matrimonial Strategist, 23:9, pp. 3-8 (2005).

Most importantly, Martindale notes the stealthy nature of bias. While some biases “operate consciously; some operate outside the realm of consciousness.” Id. In other words, an evaluator whose work has been contaminated by bias may be blissfully unaware of this fact.

Biased Evaluations Are Common

The late Jay Ziskin, author of the renowned treatise “Coping with Psychiatric and Psychological Testimony,” (Law and Psychology Press 1995), stated that he had never encountered a forensic report where bias was not present. Id. at p.1401-02. In many instances, Ziskin asserted that “clinician’s biases may be the only actual determining variable because almost any parent will provide some bases for rationalizing the clinician’s conclusions.” Id. at p.1402. Even if Ziskin overstates the case somewhat, it can hardly be gainsaid that bias is a frequent and stealthy visitor in the custody courtroom.

Stephen L. Golding, a noted forensic psychologist, stated that “biased advocacy is … rampant” in custody evaluations. Golding, S.L., “Mental Health Professionals and the Courts: The Ethics of Expertise,” in Bersoff, D.N., “Ethical Conflicts in Psychology,” Second Edition, (APA, 1999), p. 478.

Wittmann has also underscored the ubiquity of forensic bias:

Our field is famous for supporting conclusions during testimony simply on the basis of ‘accumulated clinical experience,’ a phrase which may mean nothing more than accumulated personal bias.

Wittmann, J.P., “Child Advocacy and the Scientific Model in Family Court: A Theory for Pre-Trial Self-Assessment,” The Journal of Psychiatry and Law, 13(1), (1985), pp. 77-78.

Professor Robyn Dawes, recipient of the APA’s William James Book Award, stated:

Such biases are particularly strong either when judgments are made in the absence of a well-validated scientific theory or when they are evaluated without systematic feedback about how good they are. Unfortunately, both of those conditions characterize the art of clinical prediction in professional psychology.

Dawes, R.M., “House of Cards,” The Free Press (1994), p.27.

As will be shown below, when measured against Dawes’ observations, the “best interests of the child” (BIC) concept is stunningly susceptible to bias. DRL, Sec. 240.

‘BIC’: Visionary Vessel of Values

It is not at all surprising that bias is a prevalent, ophidian presence in custody evaluations. This is especially so where courts allow or even demand that the evaluator opine as to the decisive question of what custodial arrangement is in the best interest of the child.

Although sometimes called a standard, “best interests” is not. It is in no way an operationally defined specification. It is at best a legal construct representing an aspiration, one laden with personal, social, cultural and moral value judgments. Significantly, there is no corresponding psychological construct in the behavioral science field. There is not even an objectively measured outcome criterion.

A quick thought experiment demonstrates this reality. Assume that one can perfectly predict how a child, let’s call him Alfred, will turn out depending upon which parent gets custody. If parent “A” gets custody Alfred will be an intensely motivated high-achiever, will excel in college and post-graduate education and will go on to tremendous professional and financial success, the fruits of which he will philanthropically share with the world, another Alfred Nobel if you will. The downside to such a hard charging life, of course, is that he will periodically suffer from stress and anxiety, perhaps punctuated by ulcer eruptions and bouts of depression that will require pharmaceutical intervention.

In contrast, if Alfred is placed with parent “B” he will barely graduate high school, will have only sporadic employment, and will spend much of his adult life on his mother’s couch, watching game shows while clutching the remote control in one hand and a bag of Cheetos in the other. In sum, he will traverse the path of life without leaving the faintest of footprints. On the upside, he will be blissfully stress-free, happy as a pig in slop, a modern-day Alfred E. Neuman.

So, which outcome is better? Which Alfred do you prefer?

Reasonable minds can differ and much of that difference will lie in the values you bring to bear on the decision. And that is the point. Whichever outcome you may choose, behavioral science research will not prove you wrong because your decision was driven, consciously or otherwise, by value judgments—just as are the decisions of custody judges—and values lie beyond the reach of behavioral research and, thus, beyond the limits of psychological expertise. Perhaps that is what one jurist had in mind when penning the refreshingly candid words:

At the bottom line, what is in the child’s best interest equals the fact finder’s best guess.

Montgomery County Dept. of Social Services v. Sanders, 38 Md. App. 406, 419 (1977).

Thus, the first condition conducive to bias articulated by Dawes, the absence of a well-structured standard or theory, is fulfilled with a vengeance in the custody arena. As to the second, the absence of systematic corrective feedback, that is also at work. In most cases, the evaluator submits the report and never hears another word about the matter because the case gets settled and there is no need to testify. Even when the evaluator testifies it is only by happenstance that he or she may later learn the outcome. Thus, the evaluator does not know whether the forensic recommendations were followed or not. Finally, to whatever extent children have been placed in accordance with an evaluator’s recommendations, there is no system in place for tracking those children to adulthood to see how well they fared. Consequently, the evaluator does not know how many of those children went on to successful lives versus how many are languishing in prisons or mental institutions.

The Myth of Objectivity

There is an unfortunate and wildly unwarranted assumption in some quarters that the fact that custody evaluations are performed by court-appointed experts, rather than retained partisans, extirpates bias from the process. Ochs v. Ochs, 193 Misc.2d 502 (Sup. Ct., Westchester, Spolzino, J. 2002). Court appointment may eliminate allegiance bias (Neal, T., Grisso, T., “The Congnitive Underpinnings of Bias in Forensic Mental Health Evaluations,” Psychology, Public Policy, and the Law, Vol. 20, No. 2 (2014) PP. 200-211), also called “retention bias,” which flows from paid partisanship, but it does not eradicate other types of bias that threaten reliability—and there are many!

Wittmann delineates no fewer than 15 species of bias that may contaminate a forensic evaluation; and that is but a partial catalog. “Evaluating Evaluations,” supra at p.160. His list includes, inter alia, affiliative bias, credibility bias, confirmatory bias, data-selection bias, instrument bias, pathology bias, primacy/recency bias, intervention bias and parenting values bias. Significantly, none of these contaminants is deracinated by the fact of court-appointment.


A variety of biases contaminates human judgment, including that of custody evaluators and the best interest paradigm is so amorphous as to be a veritable petri dish of opacity within which bias can flourish. Accordingly, lawyers and judges must be vigilant in examining the role that bias may have played in generating forensic conclusions before relying upon those conclusions to alter the life-course of the child whose custody is in issue.