*1108OPINION OF THE COURT
David O. Boehm, J.The respondents, Margaret R. Kuhn, Assessor of the Town of Webster and the Board of Assessment Review of the town, move to consolidate a series of tax certiorari proceedings dealing with tax assessments on the property of petitioner, Xerox Corporation. For each tax year from 1976 through 1986 petitioner has commenced a tax certiorari proceeding, and in each such year until 1983 the successive petitions were consolidated with the pending proceedings from prior years by application of petitioner. Respondents’ motion seeks to consolidate the proceedings for tax years 1984, 1985 and 1986 with those previously consolidated.
The petitioner cross-moves for partial summary judgment declaring the town-wide ratio for tax year 1984 to be a weighted mean value of 9.95%. Alternatively, petitioner seeks a declaration that the method by which it obtained the 9.95% value, i.e., by a comparison of the 1984 fractional assessment with the 1985 full value assessment, is a permissible method of proof under Real Property Tax Law § 720 (3).
Prior to July 1, 1985, the Real Property Tax Law allowed the use of three different methods for proving inequality of a tax assessment — use of the State equalization rate, sales ratio studies and the selected parcels method. In reliance upon the law as it then stood, petitioner in 1982 commissioned sales ratio studies using every property within the Town of Webster for the years 1976 through 1984, at a cost to it of nearly $200,000.
However, in 1985 an amendment to Real Property Tax Law § 720 (L 1985, ch 878) prohibited the use of sales ratio studies in all proceedings which had not been finally determined as of the amendment’s effective date (Aug. 2, 1985). Prior to its expiration on June 30, 1986, section 720 was again amended (L 1986, ch 679), effective July 1, 1986. The 1986 amendment continued the prohibition against the use of sales ratio studies, and modified the selected parcels method by providing for the random selection of parcels from stratified categories.
The study referred to was conducted by Dr. Allison R. Manson, a professional statistician and a professor in the Department of Statistics at North Carolina State University in Raleigh, North Carolina. In his study he calculated the degree of variation in individual assessment ratios within the town, thereby determining the number of randomly selected *1109parcels needed to produce an accurate estimate of the average assessment ratio throughout the town. Manson did this by-taking the assessed values appearing on the town’s 1985 assessment roll (the year in which the town began full value assessment) and comparing those values to the individual assessment ratios for every parcel on the 1984 assessment roll which also appeared on the 1985 roll.
Before computing the variation, Manson deleted all properties not contained in both the 1984 and 1985 assessment rolls, properties that had been subdivided into several parcels on the 1985 roll, properties of public utilities (the statute prohibits their use in the sampling process), properties whose classification had changed between 1984 and 1985, and the Xerox parcels at issue. Dividing the 1984 fractional assessment by the 1985 full value assessment, Manson found a dispersion from a low ratio of .1% to a high of 786.7%. Therefore, because of this broad dispersion, Manson calculated that use of the stratified random sample method would require a minimum of 600 parcels to be selected and appraised in order to obtain an estimate of the weighted mean ratio which would be within 5% of the actual weighted mean ratio. His calculations concluded that the weighted mean ratio between the 1985 full value assessment and the 1984 fractional assessment is 9.95%.
Real Property Tax Law § 720 (3) provides for only two methods of proving assessment ratio outside of New York City or Nassau County. One method, the State equalization rate, is not here available because petitioner’s property represents such a large percentage of the town’s total assessed valuation, and also because it was included in the sample of properties used by the State in computing its equalization rate (see, Matter of Standard Brands v Walsh, 92 Misc 2d 903, 911, affd on opn below 60 AD2d 605, lv denied 43 NY2d 649). The other method provides for the selection of parcels as determined by "stratified random sample”, and requires that the sample parcels be chosen randomly rather than selectively by the parties.
A stratified random sample is achieved by "the sorting of all taxable parcels except public utility property, on the assessment roll to be used for the selection of parcels, into a number of mutually exclusive categories each of which is sampled independently in such a manner that each parcel in each such category shall have an equal opportunity to be selected” (Real Property Tax Law § 720 [3] [a] [3]). Evidence of actual sales *1110during the assessment year in issue is not admissible. The statute does not require a minimum number of samples.
Based upon Manson’s analysis, petitioner urges that it is entitled to partial summary judgment declaring that for purposes of this proceeding the town-wide ratio for tax year 1984 is this weighted mean of 9.95%. This would not be in conflict with section 720, petitioner argues, because the statistical comparison of the assessed values for all properties on the 1984 assessment roll with their full value on the 1985 roll is consistent with the "stratified random sample” procedure taken to its extreme; whereas section 720 requires the selection of a number of parcels to be appraised, the use of the weighted mean average of 9.95% would include every parcel. This result is not merely an estimate of ratio, but is in fact the true ratio, and, to obtain a similarly accurate result, petitioner points out, it would require the costly appraisal of at least 600 parcels, thereby imposing an unfair burden upon petitioner since it has already expended nearly $200,000 in commissioning the now inadmissible sales ratio studies.
Furthermore, requiring new appraisals would be statistically less accurate, petitioner argues. The town reassessed all of the taxable parcels of property at full value, effective May 1, 1985, and it is, therefore, possible to directly calculate the true ratio in the town in 1984 by comparing the assessment of each property on the 1984 assessment roll with its full value as reflected in the 1985 roll. This method would yield the actual average ratio in 1984 and the ratio so obtained could be extrapolated to each prior year in litigation. Petitioner states it is quite willing to accept the full values as assessed by the town in 1985 and to use them in establishing the actual, rather than an estimated, ratio. Such actual ratio has already been calculated; it is 9.95%, and should be so adjudicated, petitioner urges.
Petitioner’s argument is very persuasive. Its approach seems to produce the most accurate calculation of ratio because of the number of parcels considered and its fairness is appealing because of its willingness to use the town’s own estimates of full value in establishing ratio. In addition, it spares the parties the expense of making appraisals on the number of properties required by the stratified random sample method. However, close analysis unfortunately requires rejection of what might otherwise be an expeditious and economic solution.
*1111To begin with, the accuracy of Manson’s final ratio is based upon the assumption that the market value of each parcel in 1984 was the same in 1985. Adrian Stanton, the Town Supervisor, states in his affidavit, however, that the average home selling price in the town rose an average of 7.3% from 1984 to 1985, that improvements were made on a number of the parcels as evidenced by the issuance of 357 building permits in 1984 and another 446 in 1985, that the value of some parcels was reduced by fire and demolition, and that some 1984 assessments inaccurately reflect value, having been corrected by administrative or judicial determinations in article 7 proceedings, and other article 7 proceedings for both 1984 and 1985 are still pending. Stanton also refers to the certificate from the State Board of Equalization and Assessment establishing a final State equalization rate of 155.85 for the town’s 1985 assessment roll, indicating that the assessments established by the assessor are not reflective of full value.
Regarding the selected parcels method, the statute provides: "The parties shall mutually agree on the methods of stratification, computation of ratio and selection of the parcels to be appraised, the number of such parcels and the number of witnesses to be heard with respect to such issue * * * In the event that proceedings for more than one year are consolidated for trial, evidence may be introduced to adjust the ratio derived from the selected parcels provided for in this subdivision so as to make such ratio applicable to the other assessment rolls included in such consolidation.” (Real Property Tax Law § 720 [3] [b] [1].)
The statute specifically limits and regulates the procedure by which ratio is to be proven. Adopting petitioner’s argument would, in effect, be legislating an alternative procedure. In interpreting legislation, courts must avoid engaging in judicial legislation under the guise of statutory construction.
One premise underlying petitioner’s motion is that the respondents should be bound by the values the town adopted in implementing full value assessments in 1985. Petitioner argues that the town’s adoption of the values and its representation to the public that they reflect fair market value constitute a binding admission of value which is controlling in this proceeding also.
However, a party against whom an extrajudicial admission is offered has the right to offer an explanation (Cohen v National Cas. Co., 275 App Div 1062; Richardson, Evidence *1112§§ 209, 228 [10th ed]). Further, the effect of such an admission is not conclusive. It raises a question of fact along with all of the other evidence in the case (Gangi v Fradus, 227 NY 452, 456-457; Richardson, op. cit. § 229).
Stanton’s affidavit points out a number of factual distinctions impacting on value between what may have appeared to Manson as identical parcels contained in both the 1984 and 1985 tax rolls. This raises a serious question as to the propriety of using the 1985 full value assessment as an indicator of market value at the time of the 1984 assessment. While the adoption of the 1985 assessment might be an admission of value at the time of the 1985 assessment, an issue not presented here, there are questions of fact whether it is representative of market value at the time of the 1984 assessment.
The cases cited by petitioner to support its point that respondents are bound by their own full value assessments do not mandate a contrary result. Samuels v Town of Clarkson (91 AD2d 837) simply places the primary responsibility for assessments upon the assessor who, alone, has the obligation to make the final determination after verification and independent evaluation of an independent appraiser’s figures. Matter of Trinity Place Co. v Finance Administrator of City of N. Y. (72 AD2d 274) merely recognizes that there is a presumption of validity of assessments which must be overcome by a petitioner challenging an assessment as excessive. And both Matter of Town of Oyster Bay (Bruce) (54 AD2d 762) and Niagara Falls Urban Renewal Agency v Clifton Holding (43 AD2d 900) involve the pretrial disclosure of appraisals. None of the cases involve appraisals being used as an admission against a party to conclusively establish value.
Petitioner attempts to diminish the distinctions raised by Stanton by pointing out that the 1985 assessment roll was completed less than seven months after the 1984 assessments had been adopted, that such distinctions are insignificant considering that the 1985 assessment roll constitutes approximately a billion dollars in value and is made up of approximately 900 parcels, that demolitions and additions significant enough to result in a change in use were deleted by Manson and that other less significant demolitions and additions are to some extent offsetting. Petitioner also points out that use of the equalization rate is permissible even where no adjustment for appreciation of value has been made for three to five years. Lastly, if imperfections exist, petitioner argues, its method is still more accurate than the selected parcels *1113method using a smaller sample. However, given that the function of a court upon a summary judgment motion is issue finding, not issue determination (Sillman v Twentieth Century-Fox Film Corp., 3 NY2d 395), these arguments are unavailing.
Although the town’s appraisals conducted for the 1985 assessment may be close to the actual values in 1984, and the method petitioner proposes may possibly yield a more accurate result because of the number of parcels sampled, such considerations have little relevance. The issue is not whether the method petitioner proposes is accurate but rather whether it complies with section 720. There is no question but that the amendment of section 720 while these assessment proceedings were pending has nullified the expenditure by the petitioner of an appreciable sum of money, but this gives no legal support to its position.
The Legislature has the power, which it exercised in this case, to alter existing rules of evidence and establish new rules (Board of Commrs. of Excise v Merchant, 103 NY 143). A litigant possesses no vested right in a rule of evidence (Cook v Town of Nassau, 40 AD2d 1050, affd 33 NY2d 7).
Petitioner’s suggested approach is erroneous for another reason. It fails to compare the 1984 assessment with true value in that same year; rather, it compares the fractional assessment in 1984 with the full value assessment in 1985. Plainly contemplated by the selected parcels method in section 720 is a comparison of assessed value with true value at the time of the assessment (see, Matter of Wolf v Assessors of Town of Hanover, 308 NY 416, 422-423). It is not really necessary to encumber the record with a large and burdensome number of parcels in order to arrive at the proper ratio. Section 720 merely requires that the parties agree to the number of parcels to be used and if they are unable to agree the court makes that determination. All that is required is a fair cross section (see, Matter of Wolf v Assessors of Town of Hanover, supra; People ex rel. Hagy v Lewis, 280 NY 184).
Further, section 720 gives each party the right to conduct independent appraisals of the selected parcels and to present proof in depth and detail as to each parcel selected. Petitioner’s contention that the method it proposes is nothing more than the selected parcels method taken to the extreme overlooks the right of the respondent to commission their own appraisals and, at trial, to subject petitioner’s appraisers to cross-examination regarding the myriad of factors which determine value.
*1114For the foregoing reasons, petitioner’s cross motion is denied.
Because of the numerous common questions of law and fact, respondents’ motion to consolidate is granted. Petitioner’s application to segregate the 1984 proceeding because of the present and future impact of that year’s assessment, which was used to determine the homestead dual rate tax structure for all future years, is denied. Having rejected petitioner’s contention that a comparison of the 1984 and 1985 assessements comports with subdivision (3), there is no basis at this time to conclude that a resolution of the 1984 proceeding could proceed any more expeditiously and without a duplication of proof if segregated from the other proceedings.