OPINION
Defendant appeals his conviction of possession of more than eight ounces of marijuana. NMSA 1978, § 30-31-23(B)(3) (Repl.Pamp.1989). Although defendant raises a number of issues on appeal, the dispositive issue is the sufficiency of the affidavit for the search warrant used in the investigation of the case. We hold the affidavit insufficient and reverse on that basis. Because double-jeopardy doctrine would prohibit a new trial if the evidence admitted at trial had been insufficient to *262sustain a conviction, we also address the sufficiency of the evidence.
We set forth the affidavit in its entirety:
•1. The Affaint [sic], Luis Lara, is a full-time salaried law enforcement officer with 14¥2 yre [sic] experience and certified by the state of New Mexico. Currently serving at [sic] a Lutinent [sic] with the Chaves County Sheriffs Office.
2. On 8-1-88 Affiant received information from the Chaves County Crimes-toppers that a caller advised that, at the location discribed [sic] on the face of the search warrant, a subject known to him as Steve lives there and is growing a marijuana patch behind a barn west of the house.
3. The caller advise [sic] that he is familiar with marijuana and knows what marijuana and marijuana plants look like.
4. Affiant did follow the callers [sic] directions and located an older model yellow El Camino which caller discribed [sic] and also a red/white Ford p/u which caller discribed [sic] at the residence which also had a small shed or barn southwest of residence[.]
5. Sgt. Troy Grant advised Affiant that a Confiditual [sic] Informate [sic] which Sgt. Grant personally knows to be reliable stated that subjects at that discribed [sic] residence would be stripping the plants and moving them out before morning.
6. Affaint [sic] who has approximately 12 yrs. experience in narcatic [sic] investigation does know that people who grow marijuana and process marijuna [sic] after stripping the. plants will process, wiegh [sic], and package it inside the residence or in an enclosed area.
Based on the affidavit, a magistrate approved a search warrant for a home, including all outbuildings, appurtenances, and curtilage.
The only allegations of criminality in the affidavit are hearsay from persons who were not law-enforcement officers — a crime stoppers caller and a confidential informant. In recent decades courts have struggled to establish standards for evaluating hearsay used in search warrant affidavits. The United States Supreme Court has settled on a “totality of the circumstances” standard for evaluating affidavits. See Illinois v. Gates, 462 U.S. 213, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983). In State v. Cordova, 109 N.M. 211, 784 P.2d 30 (1989), however, the New Mexico Supreme Court has adopted, as a matter of state constitutional law, a test first formulated by the United States Supreme Court in Aguilar v. Texas, 378 U.S. 108, 84 S.Ct. 1509, 12 L.Ed.2d 723 (1964) and Spinelli v. United States, 393 U.S. 410, 89 S.Ct. 584, 21 L.Ed.2d 637 (1969), overruled in part by Illinois v. Gates.
There are two prongs to the Aguilar-Spinelli test for evaluating hearsay from informants: The first prong requires the affidavit to establish the veracity of the informant. The affidavit must establish that the informant is a credible person or that there is special reason to believe that the informant was truthful in providing the specific information to law enforcement (such as the statement’s being satisfactorily corroborated or being against the informant’s penal interest). See State v. Cordova, 109 N.M. at 213 n. 2, 784 P.2d at 32 n. 2. The second prong requires that the affidavit present facts establishing the adequacy of the informant’s basis of knowledge. The information supplied by the informant should be based either on the personal knowledge of the informant or on reliable hearsay received by the informant. Even prior to Cordova, the two-prong Aguilar-Spinelli test was imposed in New Mexico by the rule of criminal procedure governing affidavits for search warrants, SCR A 1986, 5-211, which states:
E. Probable cause. As used in this rule, “probable cause” shall be based upon substantial evidence, which may be hearsay in whole or in part, provided there is a substantial basis for believing the source of the hearsay to be credible and for believing that there is a factual basis for the information furnished.
*263The only prong of the Aguilar-Spinelli test that we need consider here is the veracity prong. The affidavit failed this prong because it did not establish either (1) that the informants were truthful persons, (2) that the informants had particular motives to be truthful about their specific allegations, or (3) that the allegations of criminality had been sufficiently corroborated.
First, we consider the statement by the confidential- informant. Sergeant Grant’s assertion that he knew the informant “to be reliable” is insufficient to establish the truthfulness of the informant. The New Mexico Supreme Court has approved an unadorned allegation that “the informant had provided information in the past which the affiant ‘did find to be true and correct from personal knowledge and investigation.’" State v. Cordova, 109 N.M. at 217, 784 P.2d at 36. But that is apparently as far as any court has gone in allowing the affiant to vouch for an informant without providing specifics. See 1 W. LaFave, Search and Seizure § 3.3(b), at 635 (2d ed. 1987). We are aware of no judicial ruling that the veracity prong of the Aguilar-Spinelli test is satisfied by an assertion that an informant is simply “known to be reliable.” See State v. Koncir, 367 So.2d 365 (La.1979); W. LaFave, supra, § 3.3(b), at 638. In Aguilar itself the affidavit stated only that the informant was a “credible person.” The United States Supreme Court concluded that the allegation was insufficient — it was a mere conclusion that provided none of the underlying circumstances from which a magistrate could make an independent determination. As Professor LaFave has written:
Aguilar and Spinelli thus make it clear that while assessments of informant credibility are ordinarily based upon past performance, a mere allegation that the informer is “credible” or “reliable” does not permit the judicial officer to assume that there was such past performance and that the performance was of a nature to merit such a characterization of the informant.
Id., § 3.3(b), at 627-28.
We also note that there is no particular reason to believe that the specific allegations by the confidential informant were truthful. The allegations were not against the informant’s penal interest. Nor did the observations of the officers corroborate the informant’s allegations concerning criminal conduct at the location to be searched. The affiant’s statement concerning his personal knowledge of the business of stripping marijuana plants, etc., may show that the informant also had some general knowledge of such operations, but it has no bearing on the credibility of the informant’s specific allegations. Therefore, the only possible corroboration would be the allegations by the crime stoppers caller, which we now address.
By what standard should we evaluate hearsay from a crime stoppers caller? We recognize that there may be sound reasons for crediting statements by persons who are victims of or witnesses to a crime. The majority rule appears to be that “if the citizen or victim informant is an eyewitness this will be enough to support probable cause even without specific corroboration of reliability.” Allison v. State, 62 Wis.2d 14, 23, 214 N.W.2d 437, 442 (1974). See United States v. Fooladi, 703 F.2d 180 (5th Cir.1983); People v. Ramey, 16 Cal.3d 263, 545 P.2d 1333, 127 Cal.Rptr. 629, cert. denied, 429 U.S. 929, 97 S.Ct. 335, 50 L.Ed.2d 299 (1976) (In Bank) (private citizens who are witnesses to or victims of a criminal act, absent some circumstance that would cast doubt upon their information, should be considered reliable); State v. Villagran, 294 Or. 404, 657 P.2d 1223 (1983) (In Banc); W. LaFave, supra, § 3.4(a), at 717-19.
Decisions upholding reliance on citizen-informers often, however, rely on the citizen’s willingness to provide his or her identity to the police. See, e.g., People v. Ramey; Commonwealth v. Atchue, 393 Mass. 343, 471 N.E.2d 91 (1984); State v. Lindquist, 295 Minn. 398, 205 N.W.2d 333 (1973); State v. Villagran; W. LaFave, supra, § 3.4(a) n. 51, at 722-23 (collecting cases). Cf. United States v. Brooks, 350 F.Supp. 1152 (E.D.Wis.1972) (concern about *264relying on information from a person who gave a name, because the person, a telephone caller, may have given “an erroneous or untraceable name”).
When a citizen-informer remains anonymous, courts view the reliability of the individual in a far different light. In State v. Chatmon, 9 Wash.App. 741, 746, 515 P.2d 530, 534 (1973), the court wrote, “[I]n the general case information supplied by an anonymous informant (as opposed to one known to the police but merely unidentified) cannot, standing alone, create probable cause.” As stated by Professor La-Fave, “The anonymity takes on even greater significance when there has not even been a face-to-face confrontation between the person giving the information and the police.” Id., § 3.4(a), at 723. See Rohrig v. State, 148 Ga.App. 869, 253 S.E.2d 253 (1979); In re Betrand, 451 Pa. 381, 303 A.2d 486 (1973). But see State v. Gerber, 90 S.D. 432, 241 N.W.2d 720 (1976) (anonymous telephone calls were apparently by “concerned citizens”).
We also note that it is appropriate to be more skeptical of “citizen-informers” whose alleged information is likely to have been obtained only through association with those in the criminal milieu. To quote Professor LaFave again, “[I]t would seem that courts should be cautious in accepting the assertion that one who apparently was present when narcotics were used or displayed is a presumptively reliable citizen-informer.” Id., § 3.4(a), at 728. But see State v. Gerber (anonymous tips regarding marijuana delivery). That caution would apply here.
We should note that our concern about crime stoppers callers is not founded on the reward offered to such informants. Indeed, if the motivation of the tipster is financial, the tipster would have an incentive to be accurate, because the reward may be dependent on results. See People v. Stevens, 98 Ill.App.3d 158, 53 Ill.Dec. 536, 423 N.E.2d 1340 (1981); W. LaFave, supra, § 3.4(a), at 716. Our concern is simply that it is impossible to determine the actual motive of one making a call to a crime stoppers program. More importantly, we see no special incentive for such a caller not to report rumor as personal knowledge.
Thus, we refuse to presume the veracity of a crime stoppers caller. See State v. Brown, 96 N.M. 10, 626 P.2d 1312 (Ct.App.1981). The veracity of such a caller must be established just as it must for any other informant.
Here, the affiant did not allege that the tipster had been reliable in the past. See State v. Hudgins, 519 So.2d 400 (La.Ct.App.1988). Nor is there anything inherent in the tip (such as that it was against penal interest) to support the conclusion that the caller was being truthful on this occasion. That leaves corroboration. A crime stoppers’ tip may be the basis of probable cause if it is corroborated sufficiently (perhaps by evidence already obtained by the police). See State v. Jones, 96 N.M. 14, 627 P.2d 409 (1981); Seltzer v. State, 489 N.E.2d 939 (Ind.1986); State v. Beatty, 770 S.W.2d 387 (Mo.Ct.App.1989).
The only confirmation of the caller’s information in this case (aside from the statement of the confidential informant, which we have already held cannot be credited) related to descriptions of the premises to be searched and of two vehicles parked on the property. The corroborated information was in no sense incriminatory. Also, it was so readily available to any member of the public that the caller’s accuracy in this regard was not probative of his accuracy regarding covert criminal activity at the location.
In short, the affidavit did not satisfy the requirements of Rule 5-211 and the requirements of the New Mexico Constitution set forth in Cordova. Our society should not permit a home to be searched based on an uncorroborated anonymous telephone call. The search was unlawful and its fruits were not admissible at defendant’s trial.
We are not in any way impugning the merits or the integrity of crime stoppers programs. The function of anonymous tips, however, is to direct police investigative activities, not to substitute for them. *265We are not imposing any restriction on law enforcement officers’ following up on crime stoppers information, pursuing leads, and ultimately conducting searches, effecting arrests, etc. The problem in this case is that there was insufficient investigative follow-up (at least as reported in the affidavit) to justify a search warrant for a residence.
Although we requested briefs on the subject from the parties, we do not address whether New Mexico should adopt a good-faith exception to the exclusionary rule or whether that exception, if adopted, would appropriately be applied in this case. The state did not argue for the exception in the district court. Therefore, we need not address it on appeal. See SCRA 1986, 12-216(A).
Finally, because of the possibility that the state may wish to retry defendant without using the fruits of the unlawful search, we must also address defendant’s argument that retrial is barred because of the insufficiency of the evidence at the original trial. In determining the sufficiency of the evidence, we consider all evidence admitted, including the wrongfully admitted evidence. See State v. Post, 109 N.M. 177, 783 P.2d 487 (Ct.App.1989). Reviewing all of the evidence admitted, we conclude that there was sufficient evidence to convict. Double-jeopardy doctrine therefore does not bar a retrial. See id.
Reversed and remanded for a new trial, excluding the unlawfully admitted evidence.
IT IS SO ORDERED.
BIVINS, C.J., and CHAVEZ, J., concur.