The question presented for our review is whether the board of tax and land appeals (the board) may grant an abatement which reduces a taxpayer’s real property assessment to a higher percentage of fair market value than the median ratio for the town as a whole. We hold that it cannot.
The plaintiffs, thirty-four condominium owners in three different developments in Gilford, appealed to the board for abatement of their 1988 property taxes. An abatement was granted and their property tax assessment was adjusted to 69 percent of fair market value. The plaintiffs joined together and filed a motion for rehearing which was denied. They now appeal pursuant to RSA 541:6, RSA 71-B:12 and RSA 76:16-a, V, arguing that the board should have granted them an abatement to 63 percent of fair market value because the board found that the median ratio of assessed value to fair market value for the town as a whole was 63 percent. For the reasons that follow we reverse and remand to the board for proceedings consistent with this opinion.
A town-wide reassessment of property was conducted in Gilford for the 1986 tax year. Tax assessments on all properties were fixed at *63100 percent of fair market value. The plaintiffs appealed their 1988 assessments to the board claiming that their condominiums had increased in value to a lesser degree than other property and, therefore, that they were being taxed at a higher percentage of fair market value than other properties in the town. The board found, as the plaintiffs argued, that the town’s median assessment for 1988 was 63 percent of fair market value and that condominiums as a class had a median assessment of 84 percent of fair market value.
Once a town has generally assessed real estate taxes at a specific percentage of fair market value, which in this case was 100 percent, and property values then fluctuate, without a change in the overall assessment, the town must use some method to equalize tax assessments to insure proportionality. The New Hampshire Department of Revenue Administration (the department) each year compiles equalization ratios for all the cities and towns in the State. This equalization ratio, when multiplied by current fair market values, will produce an assessment in line with existing assessments of other property. Rather than using the department’s equalization ratio as a means of insuring taxpayer proportionality, the board has chosen to use the median assessment ratio for the town plus or minus the “coefficient of dispersion” (COD) to determine a suitable range of assessment ratios. The town argues that there is an “inherent lack of certitude connected with the [equalization ratio]” and, consequently, there is no one specific ratio, but rather a range of proportionality ratios, that is acceptable.
The board limits the acceptable range of proportionality ratios to the town’s COD, which “is a measure to determine how far the most extreme variables are from the center of the equalized valuations averaging process.” The board found “that for 1988 a good range of valuation for the Town of Gilford would have been 10 percent plus or minus the median ratio.” The board explained that the median ratio is derived by comparing the assessments of properties that have sold to their selling price and taking the midpoint in the array. The board did not, however, include in its order specifically how it reached the 10 percent COD. Using the 10 percent COD, the board found “an acceptable range for values to fall in to be proportional is ... a range from 57 percent to 69 percent (rounded).” Based upon this finding, the board reduced the plaintiffs’ assessments to 69 percent of the fair market value of each condominium.
The plaintiffs argue that the abatement should have been granted to 63 percent, the town’s median assessment, rather than to 69 percent, the highest point in the board’s acceptable range of propor*64tionality ratios. The plaintiffs contend that the 10 percent differential results in unacceptable disproportionality in taxation violating part I, article 12 and part II, article 5 of the New Hampshire Constitution. We agree.
“[O]ur constitution mandates that all taxpayers in a town be assessed at the same proportion of [fair market value].” Public Serv. Co. of N.H. v. Town of Seabrook, 133 N.H. 365, 377, 580 A.2d 702, 709 (1990). “It is well settled that the test in an abatement case is whether the taxpayer is paying more than his [or her] proportional share of taxes.” Stevens v. City of Lebanon, 122 N.H. 29, 32, 440 A.2d 451, 453 (1982); see Milford Props., Inc. v. Town of Milford, 119 N.H. 165, 167, 400 A.2d 41, 42 (1979). The plaintiffs have the burden of proving disproportionality with respect to other property in the town by a preponderance of the evidence. Stevens, 122 N.H. at 32, 440 A.2d at 453. Findings of fact by the board are final and our review is limited to questions of law. RSA 76:16-a, V; Dartmouth Corp. of Alpha Delta v. Hanover, 115 N.H. 26, 27-28, 332 A.2d 390, 391 (1975).
To establish disproportionality, the plaintiffs must show that their assessment is higher than the general level of assessment in the town. Appeal of Town of Sunapee, 126 N.H. 214, 218, 489 A.2d 153, 155 (1985). Proving the general level of assessment is at times a difficult burden to carry. Id. For example, the burden is met by evidence of the department’s equalization ratio for the town only if the town has stipulated to the validity of, or actually uses, the equalization ratio. Id. at 218, 489 A.2d at 155-56.
In this case, the town and the board used the median ratio for the town plus or minus the COD to determine proportionality, rather than the department’s equalization ratio. The board found that the median ratio for Gilford in 1988 was 63 percent. Neither the plaintiffs nor the town dispute this finding. The town argues that the median ratio plus or minus the COD is the general level of assessment and, therefore, that abatement to any ratio in that range satisfies the constitutional requirement of proportionality. The plaintiffs argue that proportionality may only be achieved by granting an abatement to the median ratio.
When asked to consider taxpayer proportionality, we have consistently held that in order to achieve proportionality all taxpayers must be assessed at the same ratio. See, e.g., Public Serv. Co., 133 N.H. at 377, 580 A.2d at 709; Appeal of Town of Sunapee, 126 N.H. at 219, 489 A.2d at 156; Amoskeag Mfg. Co. v. Manchester, 70 *65N.H. 200, 206, 46 A. 470, 473 (1899). After the abatement was granted in this case, the plaintiffs were assessed at a level 10 percent higher than the median level of assessment in the town. Thus, the plaintiffs’ property is being taxed at a level disproportionately higher in relation to its true value than is other property in Gilford. See Berthiame v. City of Nashua, 118 N.H. 646, 647, 392 A.2d 143, 145 (1978). While we recognize that the median is merely the midpoint of all the tax ratios in the town, it is representative of the general level of assessment. See Public Serv. Co., 133 N.H. at 376-78, 580 A.2d at 709.
Once the town decided to use the median ratio as the basis of its equalization process, it should have granted all abatements to that median ratio. The town has not offered any reason why these taxpayers were granted an abatement to the highest point in the range of proportionality ratios that it considered acceptable rather than to the median or to the lowest point in the range. “Again, our constitution mandates that all taxpayers in a town be assessed at the same proportion of [fair market value].” Public Serv. Co., 133 N.H. at 377, 580 A.2d at 709. We therefore hold that the board should have granted abatements to the exact ratio that was used to equalize assessments in the town.
Reversed and remanded.
All concurred.