Justice HECHT delivered the opinion of the Court.

Catherine Anne Palmore, Erin Kupcunas Faseler, Special Prosecution Unit, Sean Matthew Fifield, Special Prosecution Unit–Civil Division, Huntsville, TX, Melinda Mayo Fletcher, Special Prosecution Unit, Amarillo, TX, for The State of Texas. Andrea La Shawn Medley, IRS, Milwaukee, WI, George W. Lang II, Kenneth D. Nash, Mary Elliott McKetta, State Counsel for Offenders' Office, Huntsville, TX, Rebecca Rae Malek, Dallas County Public Defender's Office, Dallas, TX, for Michael Bohannan.
Justice HECHT delivered the opinion of the Court.

We consider in this case what qualifications an expert must have to testify regarding whether a person is a sexually violent predator and therefore subject to civil commitment for outpatient treatment and supervision. Like the court of appeals, though for different reasons, we conclude that the exclusion of expert testimony in this case requires a new trial.

I

A

The Texas Civil Commitment of Sexually Violent Predators Act of 1999 defines a sexually violent predator (“SVP”) as “a repeat sexually violent offender [who] suffers from a behavioral abnormality that makes the person likely to engage in a predatory act of sexual violence.” The Act defines “behavioral abnormality” as “a congenital or acquired condition that, by affecting a person's emotional or volitional capacity, predisposes the person to commit a sexually violent offense, to the extent that the person becomes a menace to the health and safety of another person.” A person found to be an SVP in a civil court proceeding must be ordered committed to outpatient treatment and supervision.

Before the State files suit, a person must be administratively determined to be an SVP. The Act requires that determination to be informed by an expert's “clinical assessment based on testing for psychopathy, a clinical interview, and other appropriate assessments and techniques”. Once suit is filed, the Act gives both the State and the person the right to a further expert examination. If the person is indigent, and the trial court determines that expert services are necessary, the court must appoint an expert and approve reasonablecompensation to be paid by the State.

The Act does not prescribe the qualifications for experts to testify whether a person has the behavioral abnormality required for an SVP. It does provide that “[a] person who suffers from a behavioral abnormality as determined under this chapter is not because of that abnormality a person of unsound mind for purposes of Section 15–a, Article I, Texas Constitution.” Section 15–a provides in part that “[n]o person shall be committed as a person of unsound mind except on competent medical or psychiatric testimony.” Thus, in the Legislature's view, an expert used to assess whether a person is an SVP is not constitutionally required to be a physician.

The State must bring SVP commitment proceedings in “a Montgomery County district court other than a family district court”. There are seven district courts in Montgomery County. One is required to give preference to family cases. One other, the 435th District Court, is required to give preference to SVP commitment proceedings. At present, most such proceedings are assigned to that court.

B

In September 1982, Michael Wayne Bohannan, then 26, married, and employed as a machinist, rode his bicycle past K.C.'s home several times and watched her inside through a window. One evening, he donned a ski mask and carrying a large knife, entered the home through the rear door, walked down the hallway past a room in which a child was sleeping, and entered K.C.'s bedroom. She was lying on the bed, reading a newspaper. Bohannan forced her to perform oral and vaginal sex, then left. Looking back on it, Bohannan testified in this case that he thought he would get some satisfaction or self-fulfillment out of raping K.C., and in some way, he expected K.C., 27, to “like being raped”.

Some three weeks later, Bohannan was driving around on his lunch break when he saw P.H., 27, enter her home. He stopped, put on his ski mask, picked up his knife, and walked through the front door. P.H. was with a group of children, whom Bohannan made her move to another room. He then took P.H. to her bedroom and forced her to perform oral and vaginal sex. He now recalls, as before, he thought he would get some satisfaction from raping P.H., “maybe ... feel more of a man”.

Bohannan was apprehended and in 1983 pleaded guilty to two counts of aggravated rape with a deadly weapon, and was sentenced to 25 years' imprisonment. Court papers suggest that he committed a third rape for which he was not charged, but Bohannan denies it.

In 1991, Bohannan was released on mandatory supervision. In April 1992, he was charged with attempting to kidnap a nine-year-old girl in a K–Mart, and in February 1993, he pleaded guilty, though he now denies he committed any crime. His mandatory supervision was revoked and he was returned to prison.

In 1998, Bohannan was again released on mandatory supervision, and in 2000 he moved to South Carolina to live with his mother. While there, he was convicted of exposing his genitals to an eight-year-old girl in a toy store and sentenced to three years' imprisonment. Bohannan denies that the allegations were true. In 2002, he was returned to prison in Texas.

In 2004, Bohannan was released on mandatory supervision a third time. But in 2006, his release was again revoked, this time for viewing child pornography on a computer in a county law library. He was enrolled in sex offender therapy at the time. Bohannan denies that the charges were true.

Bohannan testified that he now knows that the rapes were wrong and devastating to the lives affected.

C

After receiving from the Texas Department of Criminal Justice a psychologist's report that Bohannan is an SVP, the State petitioned for his commitment. The State designated two experts to testify at trial: Dr. Jack Randall Price, a board-certified forensic psychologist, and Dr. Michael R. Arambula, a board-certified forensic psychiatrist. Bohannan designated Dr. Anna Shursen, whose doctoral degree is in family sciences and family therapy. Shursen is licensed in Texas as a professional counselor and as a sex offender treatment provider.

Trial in the case was set for January 16, 2009, in the 435th District Court. On December 15, 2008, a visiting judge in that court refused to allow Shursen to testify as an expert in another SVP commitment proceeding, In re Dodson. Though Shursen had testified a dozen times in other such cases, the judge stated on the record that she was “not qualified to present an opinion” on whether someone is an SVP. On December 24, the State moved to exclude Shursen's testimony in this case for the same reason, having previously moved for an extension of time to file such a motion past the deadline set by the court's docket control order. On January 8, the regular judge of the 435th District Court granted an extension and ordered the motion to exclude to be heard the first day of trial. On January 20, after a jury was impaneled and sworn, the judge conducted a lengthy hearing on the State's motion, and at the end of the day, announced that he would take the matter under advisement overnight. The next morning the court granted the motion, finding that “Shursen lacks the forensic training and experience to answer the ultimate question”, viz, “whether [Bohannan] suffers from a behavioral abnormality that predisposes him to commit predatory acts of sexual violence.” Bohannan moved for a continuance for want of evidence, which the court denied, and the State proceeded to present its case.

Price testified that he entered practice in 1983 and turned to forensic work in 1990. He teaches at the University of Texas Southwestern Medical Center at Dallas, the Southern Methodist University Dedman School of Law, and Richland College, and has written extensively on professional subjects. He stated that he has made from 25 to 30 SVP assessments, and in three to five of them, found that the individuals did not have the requisite behavioral abnormality. Based on prison records and a two-hour personal interview, Price concluded that Bohannan does have a behavioral abnormality. Using the Diagnostic and Statistical Manual of Mental Disorders (“DSM”), he diagnosed Bohannan as having “paraphilia not otherwise specified”—in lay terms, sexual deviance—and “personality disorder not otherwise specified”. Price applied two actuarial tests widely used to evaluate a sexual offender's risk of recidivism: the Static–99 and the Minnesota Sex Offender Screening Technique (“MnSOST”). Price scored Bohannan a “5”—moderately high risk—on the Static–99 and a “10”—high risk—on the MnSOST.

Arambula testified that he graduated from medical school, began training in general psychiatry, and applied for a license to practice in the late 1980s. He teaches at the University of Texas Health Science Center at San Antonio. He stated that he has done 16 or 17 SVP assessments in the last three years, and in two of them, found that the individuals did not have a behavioral abnormality. Based on available records and a three-hour personal interview, he concluded that Bohannan does have a behavioral abnormality. He diagnosed Bohannan with “paraphilia not otherwise specified with features of pedophilia, sadism, and exhibition” and with “personality disorder not otherwise specified with features of antisocial conduct”. Arambula did not use actuarial tests in evaluating Bohannan but concluded that he was at a high risk of reoffending in part because his sexual misconduct has continued since the rapes, despite the therapy he has received, and has involved children.

Shursen testified outside the presence of the jury that she has been in private practice since 2000, providing behavioral therapy treatment for sex offenders. She stated that she has received more than a 1,000 hours of training, sees more than 100 clients each week, and has made 18 SVP assessments. Like Price and Arambula, she reviewed all of Bohannan's records and interviewed him personally. She scored him a “5” on the Static–99, the same as Price, and an “8”, a little lower, on the MnSOST. She also determined from another diagnostic protocol, the Hare Psychopathy Checklist, that Bohannan is not psychopathic. Shursen testified that in her opinion, based on Bohannan's records, her interviews with him, the actuarial tests, and her experience in the field, Bohannan does not have a behavioral abnormality “at this time”. She was not asked to elaborate on the qualification.

The jury found that Bohannan suffers from a behavioral abnormality, as defined by statute, that predisposes him to engage in a predatory act of sexual violence. The trial court issued an order of civil commitment.

Bohannan appealed, as Dodson had in the other case in which Shursen's testimony had been excluded. The court of appeals decided Dodson first. It held that the statutory definition of behavioral abnormality has two separate components: an acquired or congenital condition, and a predisposition to commit a sexually violent offense. The second element, the court noted, was also part of the definition of an SVP—someone “likely to engage in a predatory act of sexual violence.” The court concluded in Dodson that while medical evidence is necessary to prove the existence of a condition, it is not necessary to prove a predisposition to sexually violent behavior. Someone like Shursen, trained in applying actuarial tests evaluating the risk of recidivism, and experienced in recognizing that risk among her patients, is qualified to testify whether a person satisfied that part of the definition of behavioral abnormality. The court determined that the exclusion of Shursen's testimony was harmful error and remanded the case for a new trial. In the present case, the court reached the same result for the same reasons.

We granted the State's petition for review.

II

To begin with, we do not agree with the court of appeals' bisection of the statutory definition of behavioral abnormality. The definition, again, is this:

“Behavioral abnormality” means a congenital or acquired condition that, by affecting a person's emotional or volitional capacity, predisposes the person to commit a sexually violent offense, to the extent that the person becomes a menace to the health and safety of another person.
Boiling it down, a behavioral abnormality is “a ... condition that ... predisposes” sexually violent conduct. The modifier, “predisposes”, qualifies and describes “condition”. The required condition is the predisposition. The condition has no other qualities, other than that it can be congenital or acquired. The condition and predispositionare one and the same. The definition might more clearly be written:

“Behavioral abnormality” means a congenital or acquired predisposition, due to one's emotional or volitional capacity, to commit a sexually violent offense, to the extent that the person becomes a menace to the health and safety of another person.
The condition and predisposition cannot be separate things, as the court of appeals tried to make them.

The concern regarding the predisposition is, of course, the heightened risk of offense. That concern is reiterated in the statutory definition of an SVP as “a repeat sexually violent offender [who] suffers from a behavioral abnormality that makes the person likely to engage in a predatory act of sexual violence.” The court of appeals in Dodson took this reiteration as further indication that the predisposition (or risk) and the condition required for a behavioral abnormality are separate things. We think a careful analysis of the definition of an SVP does not support the court's position.

Substituting the definition of “behavioral abnormality” for that term in the definition of an SVP yields this:

A person is a sexually violent predator ... if the person ... suffers from [a congenital or acquired condition

{Qualifier A} that, by affecting a person's emotional or volitional capacity, predisposes the person to commit a sexually violent offense, to the extent that the person becomes a menace to the health and safety of another person]

{Qualifier B} that makes the person likely to engage in a predatory act of sexual violence.
We see only two possible differences in the two qualifiers. One is that a predisposition to misconduct may not make a person likely to engage in it, or vice versa, that a person may be likely to engage in misconduct though not predisposed to do so. We think the import of predisposition and likelihood is exactly the same: increased risk. An increased likelihood of misconduct indicates a predisposition, and a predisposition threatens increased likelihood. In this regard, the two qualifiers are the same. The other possible difference between them is that a menacing offender may not be predatory, or vice versa, that a predator may not be a menace. If such a distinction were possible, and we do not think it is, nothing in the Act suggests that it is intended. Thus, Qualifier B simply explains or restates Qualifier A. Its inclusion in the definition of an SVP certainly does not suggest that Qualifier A is a separate element of the definition of behavioral abnormality.

Accordingly, we conclude that whether a person “suffers from a behavioral abnormality that makes the person likely to engage in a predatory act of sexual violence” is a single, unified issue.

III

As already noted, the Act does not prescribe the qualifications an expert must have to opine on whether a person is an SVP, and it contains only two provisions relevant to the issue.

One is the statement that “[a] person who suffers from a behavioral abnormality ... is not because of that abnormality a person of unsound mind”. The obvious purpose of this provision, entitled “Certain Expert Testimony Not Required”, is to express the Legislature's view that medical and psychiatric testimony, a constitutional prerequisite for committing a person of unsound mind, is unnecessary in SVP proceedings. The statutory provision was added to the Act after, and perhaps in response to, the court of appeals' decision to the contrary in an early case. Bohannan argues that the statute overrules the court's decision, but while the Legislature may certainly state its view of a constitutional provision, an authoritative construction is a matter for the courts. The State does not address the issue but obviously takes the position by its trial conduct—relying on the opinion of a psychologist—that irrespective of whether the constitution requires medical evidence for commitment of an SVP, it does not render other expert testimony inadmissible. Thus, both parties agree that the constitution does not preclude evidence in an SVP commitment proceeding from an expert who is not a physician. Accordingly, we assume without deciding that the constitution is no impediment to the admission of non-medical expert testimony in an SVP commitment proceeding.

The other provision of the Act relevant to expert qualifications states that at the administrative stage, an “expert shall make a clinical assessment based on testing for psychopathy, a clinical interview, and other appropriate assessments and techniques”. Obviously, the expert must have the training and experience necessary to perform as required—to test for psychopathy, to conduct a clinical interview, and to employ other appropriate evaluative procedures. The Act gives no reason to think that the qualifications of trial experts should be any different.

Absent more specific statutory direction, we apply the general rule, which is that an expert must be qualified “by knowledge, skill, experience, training, or education” to “assist the trier of fact to understand the evidence or to determine a fact in issue”. That a witness has knowledge, skill, expertise, or training does not necessarily mean that the witness can assist the trier-of-fact. Expert testimony assists the trier-of-fact when the expert's knowledge and experience on a relevant issue are beyond that of the average juror and the testimony helps the trier-of-fact understand the evidence or determine a fact issue.

Credentials are important, but credentials alone do not qualify an expert to testify. We have observed, for example, that “a medical license does not automatically qualify the holder ‘to testify as an expert on every medical question.’ ” “Trial courts must ‘ensur[e] that those who purport to be experts truly have expertise concerning the actual subject about which they are offering an opinion.’ ” The test is “whether ‘the offering party [has] establish[ed] that the expert has “knowledge, skill, experience, training, or education” regarding the specific issue before the court which would qualify the expert to give an opinion on that particular subject.’ ”

In evaluating an expert's qualifications, it is important to keep in mind that expert testimony must be relevant and reliable. “To be relevant, the proposed testimony must be sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute.” Determining whether an expert's theory or technique is reliable requires consideration of all pertinent factors, including

(1) the extent to which the theory has been or can be tested;

(2) the extent to which the technique relies upon the subjective interpretation of the expert;

(3) whether the theory has been subjected to peer review and/or publication;

(4) the technique's potential rate of error;

(5) whether the underlying theory or technique has been generally accepted as valid by the relevant scientific community; and

(6) the non-judicial uses which have been made of the theory or technique.
These factors may be difficult to apply to an opinion that is based heavily on an expert's individual skill, experience, or training. In no case, however, may expert testimony be admitted when “ ‘there is simply too great an analytical gap between the data and the opinion proffered.’ ”

In SVP commitment proceedings, the only fact issue to be resolved by the trier-of-fact is whether a person has the behavioral abnormality required for an SVP. The approach of all three experts in this case was consistent with the little guidance provided by the Act. All agreed that in assessing whether a person has the behavioral abnormality for an SVP, all available information should be considered, and that the person should be interviewed. They also suggest that a medical diagnosis should be made and actuarial risk tests should be applied. Still, “[o]pinions about behavior ... and psychology depend largely on the subjective interpretation of the expert”, and opinions “too dependent upon [an expert's] subjective guesswork” must be excluded. The expert's experience, knowledge, and training are crucial in determining whether the expert's opinions are admissible.

A medical diagnosis of a person's mental health may certainly inform an assessment of whether he has an SVP's behavioral abnormality, but the principal issue in a commitment proceeding is not a person's mental health but whether he is predisposed to sexually violent conduct. And in deciding that issue, as this case makes clear, the kind of evaluation done by a psychologist may be at least as important as a medical diagnosis. The usefulness of the expert's opinion in assisting the trier-of-fact rests not on the type of license the expert holds but on the expert's knowledge, training, and experience in dealing with sexual offenders. A person's training and experience in clinical interviews and actuarial tests is no less helpful merely because the person is not licensed as a psychologist.

We see no basis for limiting the experts who can testify in SVP commitment proceedings to physicians and psychologists. The purpose of the Act is to reduce the risk of those who are behaviorally predisposed to sexually violent conduct. An expert thoroughly acquainted with sexual offenders' behavior may be qualified to assess the risk a person poses to others. Accordingly, we hold that a person is not disqualified from testifying as an expert in an SVP commitment proceeding merely because the person is not licensed as a physician or psychologist.

IV

We turn finally to whether the exclusion of Shursen's testimony was harmful error.

Though Shursen lacks medical or psychological training, she has broad experience in dealing with sexual offenders. She is licensed by the State to provide treatment for sexual offenders, and while that license may not in itself qualify her as an expert on SVPs, she has also spent more than 1,000 hours in that work to obtain her license and continues to see more than a 100 patients a week. She testified that she receives continuing instruction, some of which is geared to applying and interpreting actuarial tests and to commitment assessments. She is qualified to apply such tests and has testified repeatedly in commitment proceedings.

The State argues that Shursen is qualified to assess the risk that a person will engage in sexually violent conduct only for the purpose of providing treatment, and not for the purpose of commitment proceedings. The State notes that by statute, experts on incompetency and insanity in criminal proceedings must be licensed physicians or psychologists. But the absence of any such statutory requirement in SVP commitment proceedings tends to contradict the State's argument. While the Legislature has discretion to set requirements for expert testimony on specific issues, when the Legislature has not done so, the State's attempted distinction between risk assessments for purposes of treatment and those for purposes of civil commitment is unfounded.

Risk assessments are to a degree subjective, and in evaluating an expert's qualifications to make them, it is important to know what training and experience an expert has in minimizing that subjectivity. Here, Shursen used the same well-recognized actuarial tests that Price used and was prepared to defend her scoring, based on Bohannan's records and her interview with him. Furthermore, Arambula testified that the risk assessments he uses in SVP proceedings are the same as those he uses in treating sex offenders.

The trial court's discretion in determining whether an expert is qualified to testify on a matter is broad but not unbounded. A trial court abuses its discretion when it excludes relevant and reliable evidence. Most SVP commitment proceedings are conducted in a single district court before a few judges, and appeals are almost always heard by one court of appeals. From the record before us, the trial court appears to have determined that Shursen is unqualified to testify, not for reasons peculiar to her experience and training, but because she is not a physician or psychologist. In this context, we think a more careful review of the trial court's ruling is warranted.

We conclude that the trial court abused its discretion in excluding Shursen's testimony. The State does not challenge the court of appeals' holding that the exclusion was harmful, and therefore we do not consider the issue.

* * *

For the reasons given, the judgment of the court of appeals is

Affirmed.