The primary issues we confront in this appeal from a judgment and sentence for kidnapping and sexual abuse is whether the district court erred at trial in admitting rebuttal evidence that an investigating officer failed to disclose to the defendant during a deposition, and in admitting evidence of a DNA match between the defendant’s DNA and the crime-scene DNA together with a random match probability of one in 431 billion. The court of appeals determined the district court abused its discretion in admitting the rebuttal testimony and reversed the conviction after concluding the evidence was highly prejudicial to the defendant. On our review, we vacate the decision of the court of appeals and affirm the decision of the district court as modified in this opinion.
I. Background Facts and Proceedings.
On July 3, 1998, Tanya Ross accompanied several girlfriends to the Lake Okobo-ji area to celebrate the Independence Day holiday. This was Tanya’s first visit to the Okoboji area, and she was unfamiliar with her surroundings. The girls checked into a motel located on the east side of West Okoboji called “The Inn” at approximately 6:00 p.m. The college-aged girls took turns in the bathroom getting ready to go out for the evening. They also began to consume alcoholic beverages, and other girls began to arrive at the motel room. Tanya consumed a couple of alcoholic beverages while at the motel.
Eventually, everyone left the motel room to go to Arnold’s Park, an amusement park located on the south side of West Okoboji. The group of girls left in two cars. However, Tanya somehow was left behind.
After waiting, to no avail, for her friends to return to the motel, Tanya decided to walk to Arnold’s Park. At approximately 11:00 or 11:30 p.m., Tanya started walking in the direction where she believed Arnold’s Park was located. Tanya had been walking for about fifteen minutes when a young man stopped and offered her a ride into town. The man appeared “harmless to her,” so she accepted the ride. They *791 engaged in a brief conversation. The man asked for her name and where she lived. Although Tanya did not ask the man for his name, he did tell her he was from Okoboji.
When they reached what appeared to be the town, Tanya requested to be dropped off. Instead of letting Tanya out of the car, the man accelerated speed and locked the doors with the automatic power locks. After parking in a wooded area near a lake, the man obtained a gun from the trunk, and threatened to kill Tanya if she attempted to leave. The man then raped Tanya at gunpoint. When the man completed the sex act, he left Tanya in the woods by herself.
Tanya then started walking down the road in search of help. After she had been walking for some time, a man named Heath Richter passed by and offered her assistance. As Heath was driving Tanya to the police department, he saw Officer Dan Schaffer. Officer Schaffer interviewed Tanya and called her an ambulance. These events took place at approximately 12:50 a.m. At. the hospital, medical personnel examined Tanya and obtained evidence fоr subsequent deoxyribonucleic acid (DNA) analysis.
Tanya provided investigators detailed descriptions of the assailant, the car, and the gun. She described the assailant as a white male, approximately twenty-one years old, thin build, short — approximately 5' 8", light brown hair, clean-shaven, a little acne, wearing gold-framed eyeglasses. He was wearing a green t-shirt on the night of the assault. Tanya described the car as a white four-door mid-1990’s model with red interior and automatic door locks. She believed the gun was a semiautomatic with a blue or dark-colored finish.
On August 18, 1998, Tanya identified Lincoln Belken from a photographic array as the assailant. The photograph of Belken matched the detailed description Tanya had provided to the authorities. On a previous date, Tanya had viewed a photographic array that did not include Belken for one hour and was unable to find the assailant. However, Tanya’s identification of Belken on August 18 was instantaneous. Tanya subsequently identified Belken’s white Achieva from a photograph as the car driven by the assailant on July 3.
A search warrant was executed at Belken’s parents’ residence in Spirit Lake, where he was residing at the time of the assault. A gun resembling the one described by Tanya was seized and later identified by Tanya. Several green t-shirts were also seized. Based on the positive identifications and the items seized, Belken was charged with first-degree kidnapping and second-degree sexual abuse.
On May 8, 1999, Tanya believed she saw Belken at her place of employment in Fort Dodge, Iowa. Tanyа had been working as a bartender at Applebee’s restaurant all day until 10:00 p.m. The young man she believed to be Belken was wearing a baseball cap low on his head. Tanya also claimed she saw a white Achieva in the parking lot when she left Applebee’s at 10:30 p.m. However, at a bond review hearing held a few days later, Belken presented evidence in the form of receipts, canceled checks, and witnesses that on May 8 he was in Sioux Falls, South Dakota, with his wife and daughter.
Belken filed a pretrial motion requesting the State to produce all written or oral statements made by anyone interviewed by the State during the investigation of the July 3 assault. The district court granted the motion, and ordered the State to produce all statements taken by its investigating officers. Thе district court also grant *792 ed Belken’s request for additional money for a private investigator to obtain voluntary statements from witnesses not listed in the minutes of testimony.
Trial commenced on November 17, 1999. State witnesses included Terry Klooster, an Iowa Department of Criminal Investigation (DCI) agent, and Thomas Loebach, a deputy police officer. Both of these officers were involved in the investigation of Tanya’s assault.
The State also introduced expert testimony of the DNA analyses performed on seminal fluid found on the undergarments worn by Tanya on July 3 and DNA samples obtained from Tanya and Belken. Maria Sides, a DCI criminalist in the state DNA analysis laboratory, testified Belken was a potential donor of the seminal fluid found on Tanya’s undergarments, as were one in 1000 Caucasians. The method used by Sides to analyze the DNA is known as polymerase chain reaction (PCR), and Sides used a 310 analyzer manufactured by Perkin-Elmer to conduct her analysis. The kit provided by the manufacturer permitted Sides to use six genetic sites in analyzing the DNA. The manufacturer also provided a data base to assist in generating the random match probability of one in 1000 Caucasians. Sides limited the probability match to Caucasians because Belken is Caucasian.
Jami Harman, the scientific director of a private forensic paternity and DNA testing laboratory in St. Louis, Missouri, named Genetic Technologies, Inc., was the State’s other expert witness. The district court denied Belken’s motion to exclude Har-man’s testimony, finding she was qualified to testify as an expert witness. Harman detailed her experiences and education in DNA testing and the process she conducted in analyzing the DNA and reaching the statistical probabilities in this specific case. Like Sides, Harman used a 310 analyzer manufactured by Perkin-Elmer, and relied on the data base provided by the manufacturer to obtain a random match probability. However, Harman used a profiler kit that permitted her to look at nine genetic sites. Harman concluded the DNA from the seminal fluid found on Tanya’s undergarments matched Belken’s genetic profile. Harman further testified that the probability that another Caucasian would match Belken’s genetic profile was one in approximately 431 billion Caucasians.
Belken presented an alibi defense. Defense witnesses testified Belken was in a boat on the lake with his wife (then-fian-cée) Jennifer, his parents Barry and Kari Belken, and his relatives Peter, Vickie, Rachaelle, and Aaron Larsen, to watch the fireworks display on the evening of July 3, 1998. The boat ride ended around 10:30 p.m., and Belken, Jennifer, and Aaron then went to the Belken residence in Spirit Lake. The other occupants of the boat arrived at the residence between midnight and 1:00 a.m. Belken was taking a shower at the residence at the time.
The only alibi witness that accounted for Belken’s whereabouts during the period of time when the assault occurred was Jennifer. Aaron did not testify at the trial. Jennifer testified that after she, Belken, and Aaron left the boat landing, Belken and Aaron went to the restroom at the park, and they then drove directly to Belken’s parents’ home, which was located five miles from the park.
The State attempted to discredit Jennifer’s testimony with its theory that Jennifer was providing false testimony to protect her husband. On cross-examination, Jennifer admitted she and Belken had an argument on July 3, but denied the State’s suggestion that the reason for the argument was Belken’s desire to go to Arnold’s Park that evening instead of on the boat with his family. Jennifer also refuted the *793 State’s assertion that the argument resumed when they reached Belken’s parents’ home and that Belken subsequently left her alone for a period of time. In addition, Jennifer denied telling Wanda Garloff, her former manager at Dairy Queen, in September 1998 that she was uncertain whether Belken had not committed the offense.
The State proffered the rebuttal testimony of Wanda Garloff to impeach the statements made by Jennifеr the preceding day at trial. Because the State had not previously disclosed Garloffs testimony, the court granted Belken’s request to interrogate Garloff before the presentation of her testimony to the jury. Garloff stated that sometime around September 1998 Jennifer told her she was not sure whether Belken did or did not commit the assault against Tanya. Garloff further stated she informed Thomas Loebach of the conversation with Jennifer approximately three times, and that Terry Klooster was present on one of those occasions. Garloff testified she last talked to Loebach only one month prior to trial.
Belken moved to exclude Garloffs testimony as impeachment on a collateral issue and, alternatively, as a sanction for the State’s failure tо disclose Garloffs testimony. Neither Loebach nor Klooster mentioned Garloffs statements during their depositions. In explaining its lack of compliance, the county attorney asserted he only learned of Garloffs specific testimony two days earlier as he was preparing for the rebuttal portion of the trial. The court subsequently overruled Belken’s motion to exclude.
Although the State was permitted to present Garloff as a rebuttal witness, the testimony derived from Garloff on direct examination was confusing.
Q. Did Jennifer Belken tell you she was not certain that Lincoln Belken had not committed that acts? [sic]
A. Yes, she was.
Neither the State nor Belken attempted to clear up Garloffs ambiguous response through further questions. However, during closing arguments, the State twice mentioned Garloffs testimony as a factor to consider in judging Jennifer’s credibility. Belken did not object nor did he address these statements during his closing argument.
The jury returned guilty verdicts on both counts. The district court subsequently entered judgment and sentence on both the first-degree kidnapping charge and the second-degree sexual abuse charge.
Belken appealed. He argued the district court erred in entering judgment and sentence on the second-degree sexual abuse charge as that offense merged with the first-degree kidnapping conviction. He also contended the district court erroneously admitted Harman’s expert testimony and Garloffs rebuttal testimony. We transferred the case to the court of appeals, which reversed and remanded for a new trial, finding the district court should have еxcluded Garloffs testimony as a discovery sanction.
The State appealed, and we granted further review. The State alleges the district court did not abuse its discretion in admitting Garloffs testimony, and even if there was an abuse of discretion, the error did not prejudice Belken.
II. Scope of Review.
We review rulings on general evi-dentiary issues and the admissibility of expert testimony for an abuse of discretion.
See State v. Schutz,
We review alleged violations of the merger doctrine under Iowa Code section 701.9 (1997) for errors at law.
State v. Mulvany,
III. Rebuttal Testimony.
A. Impeachment on Collateral Issue.
Belken first argues the trial court should have excluded Garloffs rebuttal testimony as impeachment evidence on a collateral issue. Generally, the State may impeach a defense witness by introducing out-of-court oral statements inconsistent with the testimony presented by the witness at trial. Iowa R. Evid. 613;
see
7 James A. Adams & Joseph P. Weeg,
Iowa Practice
§ 613.1, at 475 (1998) [hereinafter
Iowa
Practice]. In doing so, the State may use the impeachment evidence only for the purpose of undermining the witness’ credibility, and not as substantive evidence. 7
Iowa Practice
§ 613.1, at 476; 88 C.J.S.
Trial
§ 89, at 197 (1955). Furthermore, impeachment evidence is admissible only if it addresses a non-collateral issue.
State v. Odem,
We have identified two categories of independently admissible impeachment evidence.
See Roth,
B. Exclusion Based on Nondisclosure.
The request to exclude Garloffs testimony was predicated on the failure of an investigating officer to disclose the comрlete contents of his unrecorded interview with her during his pretrial deposition. Generally, discovery sanctions are predicated on the violation of an order or rule requiring discovery.
See
23 Am.Jur.2d
Depositions and Discovery
§ 373, at 647-48 (1983). In this case, however, the discovery order required the State to produce statements by witnesses obtained by the State, and any reports or narratives by investigators which contained statements of witnesses. The statement at issue on appeal was never recorded or reduced to writing and was not included in any written report. Typically, a statement must be recorded in some manner to be discov
*795
erable under an order requiring the production of a statement.
See United States v. McClure,
The defendant in a criminal case is permitted to depose witnesses listed by the State on the trial information in the same manner and effect as in civil cases, except as otherwise limited by rule or statute. Iowa R.Crim. P. 12(1). The State is required to name each witness in the trial information who is expected to testify at trial, as well as the contents of their testimony. Iowa R.Crim. P. 5(3). Rebuttal witnesses, however, are not required to be disclosed by the State.
See
Iowa R.Crim. P. 18(3). The State is not precluded from calling witnesses to rebut testimony presented by a defendant when it fails to endorse the name of the witness on the trial information.
State v. Nelson,
Our Iowa Rules of Civil Procedure define the uses for depositions. Iowa R. Civ. P. 144. Depositions may be used at a trial or a hearing, when admissible under the rules of evidence, for a variety of purposes enumerated in the rule.
Id.
Yet, the rule does not contemplate the use of depositions for discovery, although a deposition is commonly viewed as a discovery tool.
See Osborn v. Massey-Ferguson, Inc.,
In this case, Belken claims he was denied a fair trial because the State impeached an alibi witness with testimony presented on rebuttal which an investigating officer failed to disclose in his pretrial deposition. Thus, Belken attempts to use the deposition of the investigating officer as a discovery device to pin sanctions on the State. However, there was no record made to support a finding of misconduct or bad faith on the part of the prosecutor or the investigating officer.
See State v. Dye,
Nevertheless, the State does not dispute the need for disclosure of the disputed testimony. Instead, the State asserts the district court properly denied the request to exclude the rebuttal testimony because no prejudice resulted to Belken.
In reviewing the decision of the trial court to admit Garloffs rebuttal testimony, we reverse only if Belken was substantially prejudiced by the evidence.
State v. Anderson,
In determining whether to exclude evidence as a sanction for the nondisclosure of evidence in the course of pretrial discovery, trial courts typically consider a myriad of factоrs. These factors include the reason for the nondisclosure, any resulting prejudice, the ability to cure any prejudice short of exclusion, and any other relevant circumstances.
See State v. Veal,
We have previously observed that the circumstances of the failure to disclose the disputed rebuttal evidence are limited to the contents of the deposition itself and the prosecutor’s statement that he was not aware of the statement until trial. These circumstances do not necessarily support the sanction of exclusion. Thus, we turn to the prejudice prong of the analysis. Belken claims the undisclosed rebuttal evidence undermined his entire alibi defense because the State was able to impeach his main alibi witness, and he would not have called Jennifer if he knew the State would have been able to impeach her.
In examining the presence of prejudice, we observe that the disputed testimony was offered by the State at trial only for impeachment, not as substantive evidence of the crime. Generally, the risk of prejudice from a discovery violation is reduced when the only value of the evidence is for impeachment.
Veal,
Nevertheless, we must examine the claim of prejudice in the context of its effect on Belken’s trial strategy. Undisclosed evidence that likely would have caused a defendant to materially alter the trial strategy if it had been known is more likely to be prejudicial than undisсlosed evidence that was merely cumulative. John E. Theuman, Annotation,
Exclusion of Evidence in State Criminal Action for Failure of Prosecution to Comply With Discovery Requirements as to Statements Made by Defendants or Other Nonexpert Witnesses
— Modem
Cases,
Garloffs rebuttal testimony contradicted Jennifer’s alibi testimony. Yet, if Jennifer’s alibi testimony was truthful, advance knowledge of Garloffs impeachment testimony would have only enabled Belken to investigate Garloff to expose her untruthfulness or to find а reason for the contradiction.
See Reid,
On the other hand, if Jennifer’s alibi testimony was untruthful, advance knowledge of Garloffs testimony may very well have caused Belken not to call Jennifer as a witness.
See United States v. Jefferson,
In this light, the only real prejudice that could result to Belken was if the alibi evidence was false. Yet, no defendant should be protected from that type of prejudice. The prejudice which courts guard against could have been alleviated by a continuance. See id. We conclude the court properly acted within its discretion in admitting the rebuttal testimony.
Even if the trial court had erred in admitting the rebuttal evidence, we conclude any error would have been harmless.
See State v. Griffin,
We conclude there was overwhelming evidence of Belken’s guilt introduced by the State in its case-in-chief. Two independent tests matched Belken’s DNA to the DNA from the seminal fluid found in Tanya’s undergarments. Moreover, the probable frequency that a person other than Belken would match the crime-scene DNA was astronomical. Additionally, Tanya positively identified Belken as the assailant and the car he drove as well as his clothing and the gun he used.
IV. Expert DNA Testimony.
A. Background.
We have previously examined the nature of DNA and the scientific principles that permit it to be used for forensic identifica
*798
tion.
See State v. Williams,
The methods of analyzing DNA have also advanced over the years. Today, the PCR method has emerged as the predominate method of DNA typing.
See id.
at 457. It is based on scientific principles and methods that are accepted in both forensic and non-forensic settings,
United States v. Shea,
The PCR method of typing has two components, amplification and allele identification.
Shea,
The DNA analysis at the studied sites allows laboratories to determine if a suspect’s DNA matches the crime-scene DNA. See Imwinkelried & Kaye, 76 Wash. L.Rev. at 416. A match between the two samples tends to incriminate the suspect, while a non-match normally excludes the suspeсt as a source of the DNA left at the crime site. Id.
DNA analysis in criminal cases does not end with the matching
of
a crime-scene DNA profile with a suspect’s DNA. To put the significance of a match into perspective, prosecutors also offer evidence about the probability of finding two matching profiles. This is called random match probability, which represents the expected frequency of another human matching the same DNA profile in a rele
*799
vant population class.
Shea,
The process of calculating a random match probability involves several principles of genetics and statistics, and utilizes a data base of DNA samples.
See id.
at 335-37. These principles and statistical methods are widely accepted as valid, and questions about the reliability of random match probability have generally concerned the size and selection of the data base.
See
Imwinkelried
&
Kaye, 76 Wash. L.Rev. at 456-57. Data bases of DNA samples are used because it is not practical to test the entire population to determine whether an actual match exists. However, it is always possible for error to occur by chance when a data base does not contain the entire population being studied, and adjustments are then made to account for the possibility of random error.
See Shea,
The process of PCR amplification and identification can be largely automated, which has contributed to the creation of improved data bases in recent years. David H. Kaye, Symposium, Bioethics, Bench, and Bar: Selected Arguments in Landry v. Attorney General, 40 Jurisme-trics J. 193, 202 & n. 35 (2000). PCR-based tests also make it easier to examine more loci, which generally yields a more discriminate random match probability. 3 Id. at 203 n. 38.
B. Objections to DNA Testimony.
Belken does not challenge the validity of the testing methods used by Harman or the science to support the testing results. Instead, he claims Harman was not qualified to present her testimony. Belken asserts she lacked sufficient education and experience, her laboratory was not accredited, and she failed to provide any statistical foundation for her random match probability opinion.
1. Qualifications.
Like all expert testimony, the admissibility of DNA evidence must satisfy our accepted standards for admission of expert testimony. Yet, we have traditionally adhered to a liberal rule when reviewing claims cоncerning the admissibility of expert testimony.
Mensink v. Am. Grain,
*800
Any special “knowledge, skill, experience, training, or education” can qualify a witness as an expert. Iowa R. Evid. 702. However, no precise rule governs how a witness may acquire the requisite qualifications. 32 C.J.S.
Evidence
§ 525, at 329 (1996). The witness need not be considered a specialist in the particular field as long as the testimony is within the general area of expertise.
Hunter v. Bd. of Trs. of Broadlawns Med. Ctr.,
We conclude the trial court did not abuse its discretion in admitting Har-man’s testimony regarding the DNA analysis she performed for this case. Although Harman did not obtain a special license or graduate degree, she received a bachelor’s of science degree from the University of Missouri and took graduate courses focusing on DNA and stаtistical analysis from the Federal Bureau of Investigation Academy. Moreover, Harman received ongoing training in the field of serology, studying blood and body fluids, while employed as a forensic scientist by the St. Louis County Police Department for thirteen years. In addition, Harman attended numerous intensive seminars at a variety of institutions throughout the country involving DNA typing. Harman had been testing DNA samples for five years before she and her husband formed their own DNA testing laboratory in 1998. The evidence also reveals that Harman used a 310 genetic analyzer manufactured by Per-kin-Elmer and received extensive training from the manufacturer on the use of the analyzer. Furthermore, Harman has engaged in proficiency testing five times a year by the American College of Pathologists. She has never failed any оf the tests.
Considering Harman’s education, training, and experience, we conclude she was sufficiently qualified to testify as an expert witness in this case.
See Buller,
2. Random Match Probability.
Statistical analysis is important in DNA profiling, but the basis for the average match probability is based upon methods of computation that are readily accepted in the field and have been adopted by the National Academy of Sciences.
See
Imwinkelried
&
Kaye, 76 Wash. L.Rev. at 456. Computer programs are used by DNA profilers to make the calculations based on the information they have attained from their analysis, which are ac
*801
cepted and relied upon in the field of DNA profiling.
See Shea,
Although Harman is not a specialist in statistics, she has had significant training and experience in using statistical analysis in DNA profiling.
See Mercy Hosp. v. Hansen, Lind & Meyer, P.C.,
Harman was not obligated to explain the specific mathematical computations utilized during the DNA proсess.
See
Iowa Rs. Evid. 703 (bases of expert witness’ opinion testimony), 705 (expert witness need not disclose underlying data);
Mercy Hosp.,
V. Ineffective Assistance of Counsel.
Belken asserts several ineffective assistance of counsel claims, all of which we deem to be without merit. He first argues his counsel was ineffective for failing to object to the admission of Garloffs rebuttal testimony as inadmissible hearsay. Although Garloff was testifying to an out-of-court statement made by Jennifer, the State was not offering it to prove the truth of the matter asserted but for impeachment purposes only.
See
Iowa R. Evid. 801(c) (defining hearsay as an out-of-court statement by someone other than declarant offered “to prove the truth of the matter asserted”);
State v. Soivder,
*802
Additionally, Belken claims his counsel rendered ineffective assistance by not moving for a mistrial for prosecutorial misconduct and for not objecting to Gar-loffs testimony on due process grounds. Even if the State’s conceded violation of the discovery order constituted prosecuto-rial misconduct, Belken was not prejudiced or deprived of a fair trial.
State v. Anderson,
VI. Merger.
Belken argues, and the State concedes, the district court erred in entering judgment and sentence on both the first-degree kidnapping and second-degree sexual abuse charges. The State presented the abduction and sexual assault of Tanya as a single continuous occurrence.
State v. Constable,
VII. Conclusion.
We conclude the admission of Garloffs rebuttal testimony did not prejudice Belken. However, because we find the sexual abuse offense merged with the kidnapping charge, we vacate the judgment and sentence entered on the second-degree sexual abuse conviction. We otherwise affirm the district court.
DECISION OF COURT OF APPEALS VACATED; DISTRICT COURT JUDGMENT AFFIRMED AS MODIFIED.
Notes
. Although Belken does not raise this issue, we must address the State’s two attempts during closing arguments to clarify Garloff's ambiguous testimony. We must consider the comments within the context of the entire closing arguments and the evidence admitted.
See State v. Greene,
. DNA molecules are found in the chromosomes in cell nuclei.
See
Thomas M. Fleming, Annotation,
Admissibility of DNA Identification Evidence,
. The specificity of PCR-based tests depends not only on the number of loci tested, but also on the number of alleles per locus, and the allele frequencies. Kaye, 40 Jurismetrics J. at 203 n. 38. The thirteen loci selected by the Federal Bureau of Investigation for data basing, for example, "yield an average match probability of one in 180 trillion.” Id.