A jury found Defendant guilty of two counts of first degree sex offense, one count of taking indecent liberties with a child, and one count of first degree rape on 4 June 2008. The trial court entered judgment in accordance with this verdict on 9 September 2008, and sentenced Defendant to a term of 288 to 355 months imprisonment. From this judgment, Defendant appeals.
I. Facts and Procedural History
The State’s evidence presented at trial tended to show that Defendant and Amanda Biringer (“Amanda”) were married on 21 February 1998. Defendant and Amanda had one daughter, V.G., who was ten years old at the time of trial. Defendant also became the stepfather to Amanda’s son, J.B., who was fourteen years old at the time of trial.
J.B. testified at trial to the following: J.B. stated he did not like Defendant because Defendant had abused and sexually abused him on a daily basis. Defendant touched J.B. in his “private areas[,]” and Defendant made “[J.B.] put [J.B.’s] mouth on [Defendant’s] penis and put his penis in between [J.B.’s] legs and [Defendant] would try to put his penis up [J.B.’s] butt.” Defendant put his penis in J.B.’s mouth between five and ten times. Defendant would also put lotion on J.B.’s legs and simulate intercourse. Defendant always did this with J.B. in Defendant’s bedroom and when Amanda and V.G. were out of the house. Defendant sexually abused J.B. from the time J.B. was in fourth grade until he was in sixth grade. J.B. testified that Defendant tried to insert his penis into J.B.’s anus when J.B. was in fourth grade. Defendant told J.B. that if he told anyone what happened, Defendant would kill Amanda.
V.G. testified that she felt disappointed with Defendant because he raped her. V.G. described what she meant by “raped” by stating “[Defendant] placed his wrong private place in mine.” Defendant “forced [V.G.’s clothes] off’ and removed his own clothes during these times. V.G. testified Defendant committed these acts “maybe two” times over the course of approximately one year. V.G. did not tell any*117one when Defendant was abusing her because Defendant threatened to kill Amanda if she did, and V.G. believed Defendant’s threats.
Amanda and Defendant separated on 16 January 2006. On or about 10 November 2006, Amanda was going through the clothes in the backpack V.G. frequently took to visit Defendant, when Amanda and Misty Birch (“Birch”) found a pair of tom panties. Amanda asked V.G. what happened to the panties, and V.G. began to cry and then said Defendant had torn the panties. Amanda also testified that she had seen Defendant smack J.B. on the head and push J.B. down. Amanda further testified that she finally left Defendant because “it was getting too dangerous for the kids” and Defendant would not stop drinking and doing drags.
Amanda contacted Amy Stewart (“Stewart”), the Detective Sergeant over juvenile investigations at the Macon County Sheriff’s Department, after hearing what Defendant did to V.G. Stewart testified at trial that she met with Amanda, V.G., and J.B. at their home within a week of receiving Amanda’s initial phone call. Stewart first spoke with J.B., and J.B. told her that Defendant had made him “snort white powder up his nose and that it hurt his nose when he did it.” J.B. also told Stewart Defendant would make J.B. suck his penis almost every day when Amanda was not home.
Stewart also spoke to V.G., who informed Stewart that Defendant would take off all of V.G.’s clothes and remove his own clothes when no one else was home. V.G. also told Stewart that Defendant kept pictures of children in his safe, and the children were naked and crying. V.G. told Stewart that Defendant “would rub his penis on her pee-pee[,]” and that “it went jnside and that it hurt.” V.G. told Stewart that this happened approximately ten times.
Kay Kent (“Kent”), a child protective services investigator with the Buncombe County Department of Social Services (“DSS”), testified to the following: Kent received a referral on 20 November 2006 from child protective services for J.B. and V.G. Kent was required to respond within twenty-four hours, which she did by making a home visit the following day, on 21 November 2006. During her visit, Kent first interviewed V.G. using a forensic model designed not to lead the child. V.G. described the same events to Kent that she had shared with Stewart. Kent next met with J.B., whose description of Defendant’s actions was consistent with the description he provided Stewart. The forensic interview model Kent used to interview V.G. *118and J.B. is used statewide in order to gather information from children that is not leading and that looks for consistency.
After interviewing V.G. and J.B., Kent arranged for a medical examination to be conducted on the children by Dr. Cindy Brown at Mission Children’s Clinic, in Asheville, North Carolina. A child medical exam is twofold. There is another forensic interview such as the one Kent conducted and then also a medical exam in which the child is tested for sexually transmitted diseases and other physical concerns. As a result of her investigation of V.G. and J.B., Kent completed a North Carolina Case Decision Summary/Initial Case Plan, which is a mandatory part of the structured assessment case decision process. This form names all of the children and all of the caregivers involved, followed by a section in which the investigator determines whether each caregiver is substantiated as a perpetrator.
Kent testified that Defendant was substantiated as the perpetrator with regard to both V.G. and J.B. The term “substantiated” means that the examiners “found evidence throughout the course of [their] investigation to believe that the alleged abuse and neglect did occur.” In determining that Defendant was substantiated as a perpetrator, Kent and the other investigators looked at the case history involved as well as the specific allegations. Kent also conducted a global assessment which involves examining the level of supervision the children receive and whether the children’s mental needs are being met in the home.
Jerri Szlizewski (“Szlizewski”), a child forensic interviewer (“CFI”) at Mission Children’s Clinic, testified next to the following: A CFI “[interviews] children who are alleged to be abused in a nonthreatening, non-judgmental d,evelopmentally appropriate manner taking care not to lead them in any one direction.” Szlizewski interviewed J.B. and V.G. in December 2006, and the children provided information consistent with their prior interviews. During their individual interviews with Szlizewski, the children looked at girl and boy diagrams and indicated what Defendant had done to them.
Dr. Cynthia Brown (“Brown”), the Medical Director of the Child Maltreatment Evaluation Program at Mission Children’s Clinic, testified as an expert witness for the State. Brown examined J.B. in December 2006, and J.B.’s anal exam was normal. Brown testified that in cases where anal penetration had occurred, it was common to see findings “maybe five percent or less of the time.” One reason for this is that children often wait to disclose their injuries, and these *119injuries heal during that time. Mary Ormand, the nurse practitioner in the Mission Children’s Clinic, examined V.G., and Brown then reviewed the photographs taken during that examination. Brown did not observe any injuries from the pictures taken of V.G. Brown stated that in her experience and according to national reports, “very few children have findings even when there is genital to genital, penile to genital contact.”
At the close of the State’s evidence, Defendant made a motion to dismiss all of the charges, which the trial court denied. Defendant testified on his own behalf, and he denied ever physically or sexually abusing J.B. or V.G. Defendant’s mother, Catherine Ledford, and Defendant’s former landlord, Clara Ball, also testified on Defendant’s behalf. At the close of all evidence, Defendant renewed his motion to dismiss, and this motion was denied.
The juiy found Defendant guilty of first degree rape of V.G., taking indecent liberties with J.B., and two counts of first degree sex offense with J.B. Defendant renewed his motion to dismiss and made a motion for judgment notwithstanding the verdict. The trial court denied these motions. The trial court consolidated all charges for a single judgment within the presumptive range for a B-l -felony, sentencing Level II. The trial court entered judgment sentencing Defendant to a term of 288 to 355 months imprisonment, lifetime registration as a sex offender, and lifetime satellite-based monitoring. From this judgment, Defendant appeals.
II. Admission of Evidence
Defendant argues the trial court committed plain error by allowing Kent to testify that her investigation had substantiated Defendant as the perpetrator of the abuse alleged by J.B. and V.G. For the following reasons, we must agree.
Defendant failed to object to Kent’s testimony at trial, and is thus limited to plain error review. See N.C. R. App. P. 10(b)(2), 10(c)(4). In criminal trials, plain error review is available for challenges to jury instructions and evidentiary issues. Dogwood Development and Management Co., LLC v. White Oak Transport Co., Inc., 362 N.C. 191, 196, 657 S.E.2d 361, 364 (2008). “Reversal for plain error is only appropriate where the error is so fundamental that it undermines the fairness of the trial, or where it had a probable impact on the guilty verdict.” State v. Floyd, 148 N.C. App. 290, 295, 558 S.E.2d 237, 240 (2002).
*120Defendant argues that Kent’s testimony was admitted in error because it resolved the factual issue of Defendant’s guilt for the jury by expressing an opinion on J.B.’s and V.G.’s credibility. Defendant contends this case is parallel to our recent opinion in State v. Couser, 163 N.C. App. 727, 731, 594 S.E.2d 420, 423 (2004), where we held a medical expert’s opinion that the child “probably had been sexually abused” was impermissible and prejudicial because it amounted to an improper opinion on the victim’s credibility. In Couser, the defendant had been convicted of taking indecent liberties with a child and attempted rape. Id. at 729, 594 S.E.2d at 422. The only direct evidence against the defendant was the victim’s testimony and corroborative testimony from other witnesses. Id. at 731, 594 S.E.2d at 423. “There was no evidence that the victim’s behavior or symptoms following the assault were consistent with being sexually abused.” Id. The only medical evidence presented was that of abrasions which were not specific to, nor diagnostic of, sexual abuse. Id. The results of a rape suspect kit were negative, revealing “that the victim had no semen in her or on her clothing and that neither the victim nor defendant had transmitted hairs to each other.” Id.
Without the [medical expert opinion testimony], the jury . . . would have been left with only the testimony of the victim and corroborative testimony along with evidence of abrasions not necessarily caused by sexual assault. Thus, the central issue to be decided by the jury was the credibility of the victim. We conclude that the impermissible expert medical opinion evidence had a probable impact on the jury’s result because it amounted to an improper opinion on the victim’s credibility, whose testimony was the only direct evidence implicating defendant.
Unlike Couser, however, Kent was not qualified as an expert witness. Thus, Kent’s testimony did not constitute an impermissible expert opinion regarding the victims’ credibility. The State contends that Kent’s testimony merely served to corroborate the testimony of V.G. and J.B. “One of the most widely used and well-recognized methods of strengthening the credibility of a witness is by the admission of prior consistent statements.” State v. Locklear, 320 N.C. 754, 761-62, 360 S.E.2d 682, 686 (1987) (citation omitted). However, the conclusion reached by DSS was not based solely on the children’s accounts of what happened, and thus, was not merely a corroboration of their testimony. Rather, DSS conducted its own *121investigation to determine whether any of the children’s caregivers were participants in the alleged abuse. Kent described DSS’s investigation as follows:
We look at case history being involved and I was investigating these specific allegations that were reported and then I also do a global assessment. I mean I don’t just go in and ask about allegations. I ask about anything from their mental needs being met in the home, supervision. Based on all the information I gathered during the course of the investigation I never had any information to substantiate that Misty or Amanda were abusive or neglectful.
The cumulative effect of Kent’s testimony was to tell the jury that based upon a thorough investigation, DSS concluded that of the children’s three caregivers, Defendant had sexually abused them.
The dissent contends that the present case is analogous to State v. O’Hanlan, 153 N.C. App. 546, 570 S.E.2d 751 (2002), in which a law enforcement officer testified that he did not perform a more thorough investigation because the victim had survived her attack and was able to describe and identify the defendant as her attacker. Id. at 562, 570 S.E.2d at 761. This. Court held that the context in which the law enforcement officer’s testimony was given made it clear that he was not offering an opinion as to the defendant’s guilt, but rather that he was explaining why he did not conduct further scientific testing of the physical evidence. Id. Thus, even if the officer’s testimony was admitted in error, any resulting prejudice did not amount to plain error. Id. at 563, 594 S.E.2d at 762.
In the present case, however, Kent’s testimony was clearly improper, as she testified that DSS had concluded Defendant was guilty of the alleged criminal acts. Our case law has long held that a witness may not vouch for the credibility of a victim. See State v. Freeland, 316 N.C. 13, 16, 340 S.E.2d 35, 36 (1986) (harmless error where mother of victim was allowed to give opinion testimony vouching for the veracity of her daughter); State v. Teeter, 85 N.C. App. 624, 355 S.E.2d 804 (nurse who interviewed mentally retarded victim about alleged rape should not have been allowed to testify that she believed victim’s statement), appeal dismissed and cert, denied, 320 N.C. 175, 358 S.E.2d 67 (1987). Kent’s testimony that DSS had “substantiated” Defendant as the perpetrator, and that the evidence she gathered caused DSS personnel to believe that the abuse alleged by the children did occur, amounted to a statement that a State agency *122had concluded Defendant was guilty. DSS is charged with the responsibility of conducting the investigation and gathering evidence to present the allegation of abuse to the court. Although Kent was not qualified as an expert witness, Kent is a child protective services investigator for DSS, and the jury most likely gave her opinion more weight than a lay opinion. Thus, it was error to admit Kent’s testimony regarding the conclusion reached by DSS.
“In deciding whether an error by the trial court constituted plain error, ‘the appellate court must examine the entire record and determine if the . . . error had a probable impact on the jury’s finding of guilt.’ ” State v. Pullen, 163 N.C. App. 696, 701, 594 S.E.2d 248, 252 (2004) (quoting State v. Odom, 307 N.C. 655, 661, 300 S.E.2d 375, 379 (1983)). In Couser, this Court held that the improperly admitted testimony had a probable impact on the jury’s decision where the only other evidence of the defendant’s guilt was “the testimony of the victim and corroborative testimony along with evidence of abrasions not necessarily caused by sexual assault.” Couser at 731, 594 S.E.2d at 423; see also State v. Delsanto, 172 N.C. App. 42, 49, 615 S.E.2d 870, 875 (2005) (holding that admission of medical expert’s testimony that child was sexually abused by defendant in absence of any physical evidence of abuse constituted plain error); State v. Ewell, 168 N.C. App. 98, 105, 606 S.E.2d 914, 919 (holding that it was error for the trial court to allow expert testimony that it was “probable that [the child] was a victim of sexual abuse” when the testimony was not based on physical evidence or behaviors consistent with sexual abuse), disc, review denied, 359 N.C. 412, 612 S.E.2d 326 (2005); State v. Bush, 164 N.C. App. 254, 259, 595 S.E.2d 715, 718 (2004) (expert’s testimony that she diagnosed the victim as having been sexually abused by the defendant was plain error).
However, in State v. Stancil, 355 N.C. 266, 267, 559 S.E.2d 788, 789 (2002), although expert testimony that sexual abuse had in fact occurred was improperly admitted, the overwhelming evidence against the defendant led our Supreme Court to conclude “that the error committed did not cause the jury to reach a different verdict than it otherwise would have reached.” In Standi,
[although the Supreme Court did not reveal what evidence it relied upon, the prior Court of Appeals opinion in that case noted in addition to testimony of the victim and other corroborating evidence[,] there were two permissible expert opinions that the victim exhibited characteristics consistent with sexual abuse. State v. Stancil, 146 N.C. App. 234, 240, 552 S.E.2d 212, 215-16 (2001), *123 per curiam modified and aff’d, 355 N.C. 266, 559 S.E.2d 788. Further, there was evidence that the defendant had performed oral sex upon the victim and thus it was unlikely any physical evidence would have been left and that the rape suspect kit returned inconclusive. Id. Moreover, the victim in that case continued to show symptoms of having been sexually abused five days after the incident and showed intense and immediate emotional trauma after the incident. Id.
Couser, 163 N.C. App. at 730-31, 594 S.E.2d at 423. Thus, whereas the trial court erred in Stancil, that error did not rise to the level of plain error.
The evidence in the present case more closely resembles the evidence presented in Couser in that without Kent’s testimony, the jury would have been left with only the children’s testimony and the evidence corroborating their testimony. Thus, as in Couser, “the central issue to be decided by the jury was the credibility of the victim[s].” Id. at 731, 594 S.E.2d at 423. J.B. and V.G. provided detailed and consistent accounts of the sexual abuse they alleged Defendant inflicted upon them. J.B. testified that Defendant had physically and sexually abused him on a daily basis. V.G. testified that Defendant sexually abused her on two occasions over the course of a year. The children’s testimony was corroborated by the testimony of Amanda, the Detective Sergeant from Macon County Sheriff’s Department, and the child forensic interviewer from Mission Children’s Clinic. Although the children’s testimony and the corroborating testimony is strong evidence, our prior case law instructs that this alone is insufficient to survive plain error review of the testimony, of a witness vouching for the children’s credibility.
Accordingly, we are constrained by our analysis in Couser to hold it is probable that Kent’s testimony that DSS had concluded the abuse did occur and had substantiated Defendant as the perpetrator impacted the jury’s determination. We, therefore, must conclude that it was plain error to admit Kent’s testimony, and Defendant is entitled to a new trial. Because we grant Defendant a new trial, we need not address Defendant’s arguments regarding the denial of his motion to dismiss and his enrollment in satellite-based monitoring.1
*124NEW TRIAL.
Judge GEER concurs.
Judge BRYANT dissents in a separate opinion.