This is an appeal from the action of the State Board of Equalization and Assessment, hereinafter referred to as state board, ordering increases of assessed valuations. The counties involved herein, and the percentage increases, are as follows:

COUNTY RURAL URBAN-SUBURBAN

Boone 60 20

Cass 75 9

Holt 35 0

Loup 70 0

Nance 60 0

Polk 40 21

Sarpy 15 0

Wheeler 25 0

For all of the counties herein except Wheeler, the orders of the state board indicate the rural increases were predicated upon a 3-year sales assessment ratio, agricultural statistics, and the testimony of the county officials. In addition, Holt and Wheeler Counties! had appraisal studies by Justin H. Haynes & Company, and Loup County had an appraisal study by R. C. Walters Company, Inc. The order as to Wheeler County, in addition to the Haynes appraisal study, specified that it was predicated on a 2-year sales assessment ratio and agricultural statistics.

Many of the questions raised herein are discussed in *762County of Sioux v. State Board of Equalization & Assessment, ante p. 741, 178 N. W. 2d 754. Reference is hereby made to that opinion for questions raised concerning the regional hearings. Also, because of the result we reach herein, we do not discuss the constitutionality of the legislative acts more specifically referred to in that opinion. Reference is also made to that opinion for the discussion as to the use of state agricultural statistics, which were considered as to agricultural land for all of the counties involved herein. As there found, the consideration of agricultural statistics was a clear violation of the Nebraska Administrative Procedures Act, section 84-914 (3), R. R. S. 1943, set out therein. Reference is also made to that opinion for the discussion concerning the fact that the sales assessment ratio covered both the land and the improvements thereon. The rural increases made herein are for the land only and not the improvements. This creates a disparity between agricultural lands with improvements and those with no improvements, and puts an undue share of the tax burden on the latter.

Reference is made in each of the orders to the testimony of county officials. There is nothing in the testimony of the county officials of the counties involved herein which would even remotely support any of the increases made. Actually, the testimony of the county officials would in most instances definitely negative any increase for the particular county involved.

As noted in County of Gage v. State Board of Equalization & Assessment, ante p. 749, 178 N. W. 2d 759, scientific reappraisals approved by the State Tax Commissioner, completed and placed in effect within 2 or 3 years prior to 1969, were accepted and approved without change by the state board. Of the counties involved herein, Wheeler had such a reappraisal which was used for the first time in 1967. Yet, that reappraisal was ignored and Wheeler was given a 25 percent increase for its rural lands. An adjoining county, Antelope, also *763had an appraisal used for the first time in 1967, but no increase was made for that county. The 3-year sales assessment ratio for Wheeler was 22.75 percent. For Antelope it was 23.47 percent. Harlan County, with a 3-year sales assessment ratio of 23.06 and a 1967 reappraisal, was given no increase, nor was Keith County with a 23.49 3-year sales assessment ratio and a 1968 reappraisal. For some reason not apparent from the record, there is a purported appraisal study made for Wheeler which suggests an indicated ratio of 28 percent. There is no showing in this record as to why the reappraisal should be used for approximately 30 other counties where made within the last 2 or 3 years, and ignored for Wheeler.

The record indicates that a reappraisal for Sarpy County was accepted in 1966, and has been continually updated. The undisputed testimony is that the assessor adjusts after every sale, not only on the sale property but also on adjacent property. To apply an increase to this county, based on a sales assessment ratio for sales where the valuations have already been corrected on the county’s records, would double the increase as to these properties.

A comparison is made between comparable farm land along the border between Douglas and Sarpy Counties. The highest value for Douglas County was $374 per acre and the lowest was $167 per acre. The comparable land across the road in Sarpy County is assessed at considerably higher figures, so that any increase for Sarpy County would increase the value disparity between Sarpy and Douglas Counties. Because of overlapping school districts between the two counties, this would result in Sarpy County rural taxpayers bearing an undue portion of school taxes compared with the Douglas County taxpayers in those same school districts.

Cass County has overlapping school districts with Lancaster and Otoe Counties. The proposed increase for Cass County is 75 percent; for Lancaster, 29 percent; *764and for Otoe, 40 percent. For reasons set out in County of Lancaster v. State Board of Equalization & Assessment, ante p. 757, 178 N. W. 2d 772, the increase has been voided as to Lancaster County. There is a difference of only .46 percent in the 3-year sales assessment ratio between Lancaster and Cass Counties. Despite the fact that the state board order recites consideration of a 3-year sales assessment ratio based upon its conclusion as to the indicated ratio, only the year 1968 was considered.

In 1968, 49 rural sales were submitted for Loup County, but only 2 of the 49 submitted were used to arrive at a sales assessment ratio of 15 percent. Two sales are too few for sales assessment ratio purposes. What has been said as to agricultural statistics is pertinent here, because with the appraisal study it was a factor in the assessment of the increase.

Nance County had a reappraisal in 1966. The 3-year sales assessment ratio percentages are as follows:

1966 29

1967 25

1968 21

The conclusion reached by the state board as to the indicated ratio for Nance County is 21, so it is obvious that although the order suggests consideration of a 3-year sales assessment ratio, of agricultural statistics, and of the testimony of the county officials, the indicated ratio is exactly the same as the 1968 sales assessment ratio.

In Boone County the 3-year sales assessment ratio percentages are as follows:

1966 28

1967 27

1968 21

Here again, although the order refers to a 3-year sales assessment ratio, agricultural statistics, and the testimony of county officials, the indicated ratio is the same as the 1968 sales assessment ratio.

Only three of the counties, Boone, Cass, and Polk, are *765given urban and suburban increases. In all three, the orders state that the increase is predicated upon a 3-year sales assessment ratio and the testimony of the county officials. In Boone and Cass Counties, the testimony indicates that the town property is already overvalued, and that the increase would further overvalue it. The conclusion of the state board as to their indicated ratio is 26 for Boone and 29 for Cass and Polk Counties. In addition to Boone County, Thayer and Wheeler also had an indicated ratio of 26. Thayer was given a 15 percent increase and Wheeler none. Loup County, with an indicated ratio of 22, was given no increase. Grant County, with an indicated ratio of 24, was given no increase. Grant County had a reappraisal in 1966 and Loup had one in 1967. Sioux, Boyd, and Dawes Counties, all with an indicated urban ratio of 25, were given no increases. It is impossible to reconcile these discrepancies from the record.

Cass County, with an indicated ratio of 29, was given a 9 percent increase. Polk and Merrick Counties, with an indicated ratio of 29, were given a 21 percent increase. Cedar and Dixon Counties, which had had recent reappraisals but also had an indicated ratio of 29, were given no increases. Holt and Lancaster Counties, with an indicated urban ratio of 29, were given no increases.

Upon consideration of the record herein, we cannot escape the conclusion that the state board used completely different and uncorrelated methods to support its ultimate conclusions, and that no reasonable approach was made to achieve uniformity among the counties. As we said in County of Sioux v. State Board of Equalization & Assessment, ante p. 741, 178 N. W. 2d 754, it is the primary duty of the State Board of Equalization and Assessment to equalize or establish uniformity among the various counties, and, while absolute uniformity of approach may not be possible, there must be a reasonable attempt at uniformity.

Where the record of the proceedings before the State *766Board of Equalization and Assessment contains no evidence to justify an order, the action must be held to be unreasonable and arbitrary.

For the reasons given, we reverse the order of the State Board of Equalization and Assessment increasing the assessments herein.

Reversed.

White., C. J., and Carter, J., took no part in the consideration of or decision in these cases.

dissenting.

I respectfully dissent from the decision of the majority. This is an appeal from the State Board of Equalization and Assessment hereinafter referred to as Board. The counties of Boone, Cass, Holt, Loup, Nance, Polk, Sarpy, and Wheeler briefed and presented their cases together and have been so considered. The question presented is whether or not Article VIII, section 1, Consti*767tution of Nebraska, has been complied with. That section states: “Taxes shall be levied by valuation uniformly and proportionately upon all tangible property and franchises, * * Formerly, when the state levied property taxes, this provision required uniformity and equalization of assessments between counties throughout the state. With the abolition of the state property tax, uniformity between western and eastern Nebraska counties is no longer mandatory as neither is subject to taxes levied in the other. If Scotts Bluff County were assessed at 50 percent of actual value and Douglas County at 25 percent, no prejudice would result to Scotts Bluff County because there are no overlapping taxing districts between the two counties.

The record before us is absolutely barren of any evidence regarding overlapping taxing districts. I would nevertheless take judicial notice that such overlapping occurs as between adjoining counties. In the absence of such overlapping districts, any inequalities appearing in county assessments is nonprejudicial, as to such counties, or to put it another way, error without prejudice and does not merit consideration.

In preparing for the task of equalizing between counties, the Board held regional hearings at which a large volume of evidence was adduced. The transcript of this evidence was not introduced in evidence and, therefore, cannot be considered here. See Chapter 84, article 9, R. R. S. 1943. The record is also deficient in another respect. Property sales used in arriving at sales-assessment ratios were not considered at the hearing before the Board. It was understood that any party interested could go into this matter after the hearing with a representative of the Board and that information acquired would be considered. Whether or not this was done does not appear in the record. If it was done, appellants have failed to include in the record the results of these supplemental fact-finding sessions or any evidence pertaining thereto. I must, therefore, assume that the data *768upon which the Board computed sales-assessment ratios is satisfactory and correct.

Also, as part of its basic preparation, the Board employed competent appraisers to make spot checks of certain representative counties. These studies were rather exhaustive in nature and definitely tend to throw light upon the actual value of the various classes of real property within the county as compared with actual values adopted by the assessor for assessment purposes. For instance, in Boyd, Holt, Garfield, Wheeler, and Greeley Counties, three classes of irrigated land, four of dry land (cultivated), two of meadow land, and three of grazing land were considered. Two approaches to actual value were used, a market-data approach and a capitalization of income approach. Results were as follows:

Market Data Income Leased Correlated

CLASS Valuation Valuation Results

Irrigated I $ 545 $ 479 $ 480

II 400 357 360

III 235 218 215

Dry land I 400 359 360

II 270 236 240

III 185 164 165

IV 85 73 75

Meadow V 225 208 210

V-130 125 125

Grazing VI 65 60 60

VI-55 50 50

VII 35 30 30

The appraisal indicates that actual values as found by county assessors and used for assessment purposes were, without exception, considerably lower than values arrived at by the market-data and income approaches.

The record presents three different methods of determining values as a prerequisite to ascertaining if adjoining counties with overlapping taxing districts are properly equalized for taxation purposes. These methods are *769official county appraisals or reappraisals, sales-assessment ratios, and spot appraisal studies.

Boone, Cas,s, and Polk Counties are concerned with the assessment of farmlands and urban property: Holt, Loup, Nance, Sarpy, and Wheeler Counties only with farmlands. None of these counties have had official reappraisals and the other criteria were necessarily relied on in regard to them. Much is said in the record and briefs before us regarding the desirability of correlating the three different criteria of value. This cannot be done any more than it is possible to correlate value on the basis of market data and capitalization of income. Two farm or ranch units may have identical income capabilities, yet one may have a greater fair market value than the other due to location, improvements, etc. Different appraisers of the same unit will arrive at different results. Sales of property within a county may not represent a correct picture of value for a multitude of reasons. There may be an insufficient number of sales and many of them may not represent true open-market transactions. No one of the three criteria available are perfect and, as a result, an element of judgment enters into this matter and must be exercised by assessing and equalizing authorities. This factor is illustrated in the case of Boone County and its neighbors. Sales-assessment ratios for these counties show the following:

1968 Sales

Assessment Ratios Changed to

County Urban Rural Urban Rural

Boone 26 21 31.2 33.6

Greeley 25 25 29.25 29.25

Nance 32 21 32 33.6

Platte 25 21 32.25 34.86

Madison 26 21 34.84 31.92

Antelope 31 24 31 24

Wheeler 26 28 26 35

It will be noted that after adjustment by the Board, there is little material difference in assessed values be*770tween the counties with the following exceptions. Greeley still appears a bit low on the basis of sales-assessment ratios, yet on the basis of a 1969 spot study of this county, it is deemed to be equitably assessed with the others. In other words, despite the conflict with the sales-assessment approach, the study made shows the county to now be assessed at approximately 35 percent of actual value. The appraisal study indicates that land in Greeley County is valued for assessment purposes at supposed actual values ranging from 48 percent to 79 percent of true market or actual values depending on the various classifications of land. These percentages average out to 23.73 percent and a 17 percent increase brings the average to 27.76 percent. The figures are not strictly accurate as allowance has not been here made for the variation in acreages of the several land classifications. It is obvious from the appraisal study that if this were done the figure of 27.76 percent would much more nearly approach the statutory 35 percent of actual value figure and would present a minimal variance therefrom. The same is true regarding rural property in Antelope County and urban property in Wheeler County where recent official reappraisals were had. The question presented under such circumstances is that always confronted in cases of conflicting evidence. In a jury trial, the jury must determine which evidence is reliable; here the Board must make this determination. In any event, no material variations appear.

Generally speaking, with reference to the counties involved in this action, the Board appears in each instance to have relied primarily on the best evidence available and' in each instance, there is evidence in the record to sustain its action. Appellants criticize the use of appraisal' studies on the ground that L,B. 391 (Laws 1969, c. 628, p. 2528) is unconstitutional. That point does not require decision. The appraisals were made primarily on the basis set out in section- 77-112, R. R. S. 1943. That statute does not appear to exclude other criteria *771of “actual value.” It states the formula therein contained is to be used only “where applicable.” Other means, when required to determine actual value, are not excluded.

This court will not substitute its judgment for that of the State Board of Equalization and Assessment if the action of the Board is not illegal, arbitrary, and capricious. See Carpenter v. State Board of Equalization & Assessment, 178 Neb. 611, 134 N. W. 2d 272. Neither will this court interfere with the discretion of the Board when discrepancies are slight and substantial equality and uniformity are attained. County of Kimball v. State Board of Equalization & Assessment, 180 Neb. 482, 143 N. W. 2d 893. The decision of the Board as to appellant counties should be affirmed.

dissenting.

Neither the court nor the board has developed a practicable model for statewide equalization. Difficulties besetting the board are inferable from a model proposed by Pao Lun Cheng, “The Common Level of Assessment in Property Taxation,” 23 National Tax Journal 50 (1970). “As the secretary of one equalization board has said, ‘inter-county equalization is like a mule — no hope of progeny and no pride in. ancestry.’ ” Hellerstein, State and Local Taxation 139 (1969).

In this state of affairs I would try to maintain some balance between data and intuition. “But the action does not appear to have been arbitrary except in the sense in which many honest and sensible judgments are so. They express an intuition of experience which outruns analysis and sum up many unnamed and tangled impressions; impressions which may lie beneath consciousness without losing their worth.” Chicago, B. & Q. Ry. Co. v. Babcock, 204 U. S. 585, 27 S. Ct. 326, 51 L. Ed. 636 (1907) (Holmes, J.).

I would affirm the orders of the board.