The Volokh Conspiracy is a site that I don't link often, but yesterday Orin Kerr blogged on an opinion by Judge Posner, writing for the 7th Circuit Court of Appeals, "on the question of whether a government-administered psychological test is a Fourth Amendment 'search,' and if not, what it might be."

As described in the opinion [pdf], the case arose because—

Two years after Kristin Greenawalt was hired by the Indiana Department of Corrections as a research analyst, she was told that to keep her job she would have to submit to a psychological examination. The record ... is silent on the reason for so belated a demand. But she complied and later brought this suit ... against the Department and two of its officials.... She claimed that the test, which lasted two hours and inquired into details of her personal life, constituted an unreasonable search in violation of her Fourth Amendment right to be free from unreasonable searches and seizures. Also, invoking the supplemental jurisdiction of the district court, ... she claimed that whether or not the test was a search, requiring her to take it if she wanted to keep her job both invaded her privacy and deliberately inflicted emotional distress on her, and so violated Indiana’s common law of torts. She asked for damages plus an injunction that would require the defendants to expunge the results of the test from her personnel file.

So the opinion purports to inquire

whether subjecting a public employee to a probing psychological examination is a search. If it is, then it may well have been an unreasonable one in this case, and thus violate the Fourth Amendment, because Greenawalt is merely a researcher. She has no contact with prisoners, is not armed or privy to state secrets, and has no other powers or opportunities, so far as we can tell, that would warrant imposing such a condition of employment, unlike cases such as [cases involving various law-enforcement personnel, a school administrator and persons close to the White House].

Then, incredibly, at the conclusion of this paragraph Judge Posner asserts—

But we need not decide this, as we do not think a psychological test is a search.

I guess that takes care of that!

Posner draws a distinction

The opinion continues—

Almost any quest for information that involves a physical touching, which a test does not, is nowadays deemed a “search” within the meaning of the Fourth Amendment....

Posner reviews the many activities that have been deemed to constitute a search—taking a drop of blood, applying a breathalyzer test, requiring a urine sample, wiretapping, rifling an employee's desk—but concludes that a psychological test is not among them.

Nevertheless we do not think that the Fourth Amendment should be interpreted to reach the putting of questions to a person, even when the questions are skillfully designed to elicit what most people would regard as highly personal private information. The cases we have cited show, it is true, that a Fourth Amendment claim does not depend on the claimant's being able to establish an invasion of such interests that tort law traditionally protects as the interest in bodily integrity (protected by the tort of battery), in freedom of movement (protected by the tort of false imprisonment), and in property (protected by the torts of trespass and of conversion). But that is all they show, so far as bears on the issue in this case. The implications of extending the doctrine of those cases to one involving mere questioning would be strange. In a case involving sex or some other private matter, a government trial lawyer might be required to obtain a search warrant before being allowed to conduct a cross-examination--or the judge before being allowed to ask a question of the witness. Police might have to obtain search warrants or waivers before conducting routine inquiries, even of the complaining witness in a rape case, since they would be inquiring about the witness's sexual behavior. Questioning in a police inquiry or a background investigation or even a credit check would be in peril of being deemed a search of the person about whom the questions were asked. Psychological tests, widely used in a variety of sensitive employments, would be deemed forbidden by the Constitution if a judge thought them "unreasonable."

The lawyer-blogger concurs—

Although I would quibble with a few minor points, this is basically right. Being asked and having to answer questions is primarily a Fifth Amendment question, not a Fourth Amendment question.... If asking questions and getting answers were a Fourth Amendment search, the law of criminal procedure would look dramatically different than it does today.

Why is a verbal psychological test different from ordinary questioning?

I had previously thought that the forte of judges and lawyers lay in drawing distinctions, but in this case an important distinction has been obscured.

Judge Posner views a psychological test as no different from conventional interrogations—"the putting of questions to a person" or "mere questioning."1

But a verbal psychological test may be very different from ordinary questioning. To begin, all psychological tests worth their salt have two parameters associated with them—validity and reliability. (There are various types of measures of validity and reliability, so I'm simplifying here.)

Basically, validity is the measure of how well a test measures what it purports to measure. For instance, Does such and such an IQ test really measure a subject's intelligence? The answer to that question is referred to as the test's "validity" and may be associated with a number, a coefficient of correlation, that indicates how well or poorly the test results correlate with some external (extrinsic) measure of intelligence.

The second parameter—reliability—has to do with the reproducibility of the results. If I test a subject on Wednesday, will I get the same result that I got on Monday?

While lawyers, judges or police agents (or psychiatrists and psychologists) may have some tricks up their sleeves, what they don't have are studies to validate the meaningfulness of their questions nor any studies to prove their reliability. The questioner asks a question, the responder answers and the questioner may conclude whatever he likes about that answer.

Psychological tests on the other hand are designed to elicit information about the person that may be quite different from the surface meaning of the questions. To give a fairly benign (and facetious) example, "Do you enjoy outdoor sports?" might be posed on a test to determine a subject's likely compatibility with a career choice. Research may have determined that most foresters enjoy outdoor sports and that those who don't should probably look elsewhere for a career—perhaps accounting.

The test that Kristin Greenawalt was required by the State of Indiana to take is not identified in the opinion, but I suspect that it was the Minnesota Multiphasic Personality Inventory, commonly known as the MMPI (and in its current version as the MMPI-2).

This test was originally developed in an attempt to make differential diagnoses of various psychiatric disorders. Thus the results are delivered as a set of scales with such names as hypochondriasis, depression, hysteria, psychopathic deviate, paranoia, psychasthenia, schizophrenia, and hypomania. It was subsequently shown that these scales are not valid for the purposes suggested by their names, and so the modern MMPI is interpreted in a manner not obvious from the traditional nomenclature.

Richard Niolon, a psychologist writes

The MMPI 2 has been consistently ranked one of the top two psychological instruments of all psychological instruments in use by American Psychologists. It is one of the most researched tests around as well, with over 500 published articles, books, and chapters. It's not perfect, mind you, as it has both strengths and weaknesses, but used properly, it is an invaluable tool.

He also notes that the MMPI is

respected by the courts in many different areas.... The courts and their staff actually may know some of the basics about how to read the scales even.

So Kristin Greenawalt hasn't been searched?

Finally, let me mention another matter that the opinion does not touch—nonverbal psychological tests. There are a number of such tests. The most famous in the areas of personality profiling and clinical diagnosis is the Rorschach—the ink-blot test.

Would this opinion then apply to the Rorschach?

In another post yesterday, Kerr refers us to a NY Times article

New love can look for all the world like mental illness, a blend of mania, dementia and obsession that cuts people off from friends and family and prompts out-of-character behavior - compulsive phone calling, serenades, yelling from rooftops - that could almost be mistaken for psychosis.

Now for the first time, neuroscientists have produced brain scan images of this fevered activity, before it settles into the wine and roses phase of romance or the joint holiday card routines of long-term commitment.

In an analysis of the images appearing today in The Journal of Neurophysiology, researchers in New York and New Jersey argue that romantic love is a biological urge distinct from sexual arousal.

Could the State of Indiana have required a brain scan for continued employment? After all, it's not physically intrusive.